Sciencemadness Discussion Board - troubleshooting a homemade hotplate

troubleshooting a homemade hotplate

Crucible - 1-7-2010 at 17:09

I made my own explosion proof hotplate since manufactured ones are pretty expensive. I made it out of refractory cement so it would withstand high temp and pressed nichrome wire into the top, which is connected to a variac.

I fired it up for the first time last week, but it didn't get nearly as hot as I was expecting. By my calculations there should be 3.5 Amps running through the wire which ought to heat up to 1200C at that amperage. Yet after 3 hours continuous use at 100% I couldn't boil water with it.

I covered the wire over with a very thin (.25") layer of cement to seal it in place and distribute the heat more evenly. I can feel a good bit of heat escaping out the bottom, and the wire itself heats up red hot within moments. Either its losing a lot more heat than I thought through radiation or the resistance of the wire goes up significantly when it gets hot, which reduces the available amperage.

I'd be interested to hear what a few others think about it and whether I'm doomed to only boiling solvents or if I can tweak it to get more heat out of her.

100_0811.jpg - 42kB

not_important - 1-7-2010 at 18:59

Some refractory cements are fair insulators. The ones that are decent thermal conductors do so in all directions, so so far as the heating wire is concerned it is buried in stuff that it heats to a more-or-less uniform temperature; said stuff then conducts heat away in all directions.

A quick sanity check would to be to put a similar length of nichrome in series and compare its temperature with that of the buried wire - just radiated heat.

I think you would have done better to make a plate with a wide channel on top to hold the wire, then thinly coat that with low density ceramic, and finally place the wire using only small tabs of ceramic to hold the wire in place. Use a slab of aluminium for the top, perhaps with a thin ceramic wash for electrical insulation, or get some ceramic shelf paper.

JohnWW - 1-7-2010 at 19:48

Quote: Originally posted by not_important

(cut)I think you would have done better to make a plate with a wide channel on top to hold the wire, then thinly coat that with low density ceramic, and finally place the wire using only small tabs of ceramic to hold the wire in place. Use a slab of aluminium for the top, perhaps with a thin ceramic wash for electrical insulation, or get some ceramic shelf paper.

Aluminium used for such a purpose is too soft, too liable to oxidize at raised temperatures, and its melting-point (660ºC) is rather on the low side, for an hot-plate surface. Copper or stainless steel would be a much better top surface.

Contrabasso - 1-7-2010 at 22:08

You will be losing heat through all the surface, and with nichrome the resistance does go up as the temperature rises, so there will be an equilibrium where decreasing heating due to resistance rise meets increasing heat loss due to temperature rise.

Rebuild it with a bit less wire so you start at a lower resistance, so the temperature can rise further. ALSO make the insulator as a sandwich of thinner layers that way they make a better insulator. So wind the element into a 5mm slab and place two more 5mm slabs underneath that, then the heat preferentially leaves via the hot plate surface.

A 5mm thickness of aluminium is a good thermal conductor, but 5mm of multi layered aluminium foil is a good thermal insulator.

watson.fawkes - 1-7-2010 at 22:08

Quote: Originally posted by Crucible

By my calculations there should be 3.5 Amps running through the wire which ought to heat up to 1200C at that amperage.

Measure the current through the resistive element with an ammeter. A cheap ammeter suitable for the purpose will cost you less than the nichrome wire; it may not last long, but you'll get your reading.

The temperature you achieve at a given wattage thermal output is related to the thermal mass of what's being heated, the thermal resistance from heat source to heat sink (typically atmosphere), and the temperature of the heat sink. Poor insulation causes high thermal losses and lower ultimate temperatures.

peach - 2-7-2010 at 05:37

I think you're loosing too much heat through radiation / convection / conduction. And the heat generated depends on things like how thick and long the wire is.

E.g. if you've used thick bit of wire and not much of it, it won't take very much voltage at all to get 3.5 amps through it.

Changes in temperature are caused by power dissipation, so you're better of thinking of heating power than temperatures.

Power = Current x Voltage
or = I^2R
or = V^2/R

All the same thing said in a different way.

What voltage setting do you have the variac at? My hotplate is about 6" across, as a thin metal top (~a 1/8th" thick) that's supported by insulators and the element is directly in contact with the metal top. It manages 380C and uses 680W. At 3.5 amps, it'd need 194 volts across the element to produce that much heat.

I'm guessing you've worked out the voltage setting based on the resistance given for the wire per unit length.

Tell me what voltages you're using, the resistance of the wire and how much you've used.

I think the mistake you've made is to calculate the current you need based on how hot the nichrome wire becomes when it's out in the air or in a vacuum, on it's own. Wrapped in the cement, it's effectively attached to a gigantic heatsink, that's sucking all the power from the element before it can accumulate as thermal energy; a high temperature.

I'm not sure what you're using 1200C for. That is a HUGE temperature to expect from something not inside insulation (e.g. the hotplate + flask would need insulation). If you want to bake solids and things at those temperatures, you need to build a mini furnace instead. A real world hotplate will never reach that due to radiation / conduction / convection losses. You realize your borosilicate glass would actually melt at that temperature?

I'd also warn you that air pockets or moisture in the cement could cause it to pop when it gets hot. The cement will start falling apart at those temperatures and the nichrome should only be around 1050 for continuous use. Again, if you're really need 1000C type temperatures, it'll need to be a furnace of some description.

I have Dave Gingery's "Build a crucible furnace" on the desk here. He gives a recipe for DIY refractory.

4 gallons silica sand
4 gallons grog (burnt clay / pulverized firebricks)
4 gallons fire clay
3 quarts borax

Dry mix it all well, then sprinkle on water until it just barely clumps as a solid ball when squeezed. If it looks like brick mortar, that's as far as it needs to go (better dryer). Then you ram it into a form and fire it.

These guys sell some nice high temperature heating elements. And I think their site has some PDFs with information on how to best wire and install them.

One could easily make a very clean, tidy and possibly cheap high temperature furnace with one or two of these and some insulation. I'd buy it rather than make it. Refractory bricks for 1200C service are incredibly hard and don't cut particularly fast even with diamond blades or the scratch and snap method. Wools are easier to handle.

You'll notice they sell some elements with very wide bores down their centers and open ends. They'd be ideal for a tube furnace and insulating it would be easy. Slide some quartz glass tube down the middle, done.

[Edited on 2-7-2010 by peach]

[Edited on 2-7-2010 by peach]

tumadre - 2-7-2010 at 08:47

your heating element isn't going to last very long.

3.5 amps at what voltage?
verify your power levels are high enough, and there would have been significant water evaporation.

I've tried making open channels in ball clay as well, issue is that any contamination causes the clay to conduct, just enough to corrode the element, what you really want is a non porous ceramic.

peach - 2-7-2010 at 09:57

Good point tumadre, cement is relatively caustic.

The non-porous stuff you're talking about would be refractory. The higher the temperature it's made to withstand, the more aluminum oxide (ceramic) it has in it. The really pure stuff is a serious pain in the ass to shape, you can't easily cut grooves out of it.

Besides, I have no idea why you'd be trying to get 1,200C from a hotplate in the first place. It'd never going to happen.

I'd use two aluminum or steel plates, sandwich the element between the two and butter the sandwich with some none conducting insulation on either side of the element. You can bolt the two together around the edges.

If you rip the drive coil out of a brushless motor, you could spin a stir bar without any sparks from the brushes of an AC drive motor.

[Edited on 2-7-2010 by peach]

Crucible - 2-7-2010 at 15:03

The heating wire is 28 gauge resistance of 4.25 Ohms per foot. I got this to replace the 20 gauge wire I tried first, which didn't work because it was too springy and wouldn't stay in place. I forgot to adjust the length to accommodate the new higher resistance wire and ended up with more than I'd planned. I'm using just under 8 feet of wire, it measures 31 Ohms on my meter. Voltage is standard US house wiring, ie 110 + or -.

Using the power equation P=CxV I get 385 Watts of power, which does seem a little low. I could take some wire out to lower resistance, but I'm wondering if that would help or hurt. Is less wire hotter actually going to produce more absolute heat output? There is less wire after all to produce the heat. I'm thinking joules per minute here. It may be that the heat losses due to convection, radiation, and evaporation from the bath are simply greater than the total heat input into the "system" by the heating element.

To answer a few questions, the cement has a high ceramic content and is rated for 2500F. I don't actually need 1200C, I was just throwing that out there as the theoretical max. At the moment 300C should do just fine. There was a good bit of water (trapped in the cement) that oozed out the top when I first turned it on, but its been running dry ever since.

I'm not sure how to check the current while its running, so I calculated it instead. When I hooked up my multimeter while it was on, it created a short circuit and blew the fuse on the variac. Won't be trying that again!

Not sure what you mean about putting some wire in series to check heat. How would I actually measure the heat? I can see the wire is getting red/orange hot, so I don't have much concern on that score. How do you mean to hold the wire with ceramic tabs? The weight of a small tab alone won't hold it, it would have to be embedded into something. When it was wet, I held the wire in place with staples, pushed into the cement and pulled them out once it had dried. Getting the wire to stay put was the hardest part and required an extra layer on top because some bits were still coming out.

If I'm reading you right watson.fawkes then if I reduce the thermal mass, ie the water bath I'm trying to heat, then I will get higher temps. Right?

DJF90 - 2-7-2010 at 15:31

To measure current connect the multimeter in series, and for voltage, in parallel. I do hope you know what these terms mean. From the power equation P=V^2/R, recreasing the resistance at a constant voltage ought to increase your power output. Of course if you want to make calculations, you need to measure the voltage across your coil. I'm no electrical engineer; someone like woelen would be much better at helping you, but I know enough to get me by

[Edited on 2-7-2010 by DJF90]

Nicodem - 2-7-2010 at 15:36

Quote: Originally posted by Crucible

I'm not sure how to check the current while its running, so I calculated it instead. When I hooked up my multimeter while it was on, it created a short circuit and blew the fuse on the variac. Won't be trying that again!

Current is measured in series, not in paralel!

I thought that is obvious, after all it is not called electrical current for nothing.
It is the potential that is measured in paralel. Obviously, as it is an electrical potential between two different points in the circuit.

Anyway, 385W would be too low for much use already if it was that much, but that is based on the initial resistance. Most metals have a notable rise in resistance at higher temperatures (http://en.wikipedia.org/wiki/Electrical_resistance).
For example, check the resistance of an incandescent bulb, calculate what power would that give if the resistance would be the same when also when in operation (P=U^2/R) and compare that to the nominal power on the bulb. You'll see there is a huge difference, and that comes from the rise in resistivity at the couple of thousand celsiuses at which the filament heats.
For such a massive construction, with such a high area of uninsulated surface, you would need at least 1kW, as you would have to use a current regulator for most lab uses anyway, so it does not really hurt to have the max power higher than needed.

densest - 2-7-2010 at 16:03

Nichrome is unusual in that its temperature coefficient of resistance is quite low - much lower than most metals. Many nickel-containing alloys are similar.

One way to salvage the hotplate would be to dig down to the heating element at its midpoint. Connect to it there and join the original ends together. This will give you an element with 1/4 the resistance of the original. It will draw 4x the current giving 4x the heat output. 4 * 385 = 1540W which is more useful.

Single-element electric stoves are usually 600-1200W, so that's the right order of magnitude.

The points of connection between nichrome and copper are where it will be most likely to burn out. The copper will oxidize and the nichrome will get especially hot because there's nothing draining heat away from that point. Doubling the nichrome there would have been a good idea. i.e. wind an extra strand over the joint and a couple of cm into the heating section. McMaster-Carr (and other suppliers) sell nickel wire covered in glass fiber and nickel clamp connectors for this sort of application. Join the copper to the nickel at least 10cm away from the hot area.

Contrabasso - 3-7-2010 at 00:25

When making an element without access to all the nickel connector blocks and wire tails, double the element wire over for say 6" and fold the end of that double wire over again then twist it all together carefully.
This means that the free end is four thicknesses of resistance wire so doesn't get as hot, so ordinary connectors work more safely.

Get some clues about the usual resistance for heating elements by measuring the cold resistance with a meter, and also by calculating the hot resistance by power and volts relationships

P=EI
E=IR

To have some control you need an element more powerful that you think so that you can turn it down by a switching controller -which is a whole degree topic!

Think about a soldering iron! A small 15w iron will melt solder but it's SMALL, to melt lots of solder you need a bigger iron and more power simply to keep the temperature up as the surface area gets bigger.

[Edited on 3-7-2010 by Contrabasso]

peach - 3-7-2010 at 14:10

Right, that's nowhere near enough watts. My hotplate is double that and it's element is attached to a thin plate of metal, which is thermally insulated from the unit.

Think about the equation P = I^2R. The power you're producing will be exponentially more influenced by the current as opposed to the resistance, so you want to increase the current.

V = IR, so I = V/R. As you have R fixed at a certain value, you can increase I by increasing V. You'd double the current by running the element from a 220-240V supply. If you're in the US, you can get that by twinning together the two lives that are coming into your home. But I would advise extreme caution doing that if you're not used to the mains. A transformer would be a safer idea.

A shorter element will get hotter with the same voltage across it, but it won't spread the heat out very well. You could also use a thicker element.

With regards to water, you have to be concerned about water trapped inside the cement. The safest way to get rid of that is to run the plate at a low temperature for a long time to bake any moisture out.

In terms of getting this to work, I believe you need to rethink the design. Make the platter around 6" and make it thinner. Stand it on thermal insulating block from the building yard, the kind that are porous, very light and can be cut with a saw. That'll support the thinner platter and provide good thermal insulation.

A better option, in my opinion, is the one I suggested earlier. Simply get two pieces of aluminum plate (real cheap, onlinemetals.com) and sandwich the element between the two with some insulation. Drill a few holes round the edges, bolt them together. Stand that on some of those thermally insulating blocks.

I think you may struggle with the cement mix and making it thinner risks it snapping with hot chemicals on the platter.

I've often heard people going on about using low wattage soldering irons for 'careful work'. That's complete balls. If I use a lower wattage iron (15W), it cools down too much when it touches the component and takes time to heat up. During which, the heat is flowing down the leads and into the parts. I do far less damage to parts with a 30W+ iron as I can melt the solder as soon as I touch the lead, so the heat has less time to transmit down the lead. Even with some small components, I've barely been able to get them hot enough to solder with 15W irons, but the component has become roasting hot in the mean time. A lower wattage iron means more energy ends up in the component than it would with a higher wattage iron, since it needs such a long contact time, allowing conduction to take place. I've also had one of Hakko's new soldering stations, the really expensive ones. They're incredibly nice, but I can do 99% of the work with a $5, unregulated, no name iron.

[Edited on 3-7-2010 by peach]

Crucible - 5-7-2010 at 15:24

Quote: Originally posted by densest

One way to salvage the hotplate would be to dig down to the heating element at its midpoint. Connect to it there and join the original ends together. This will give you an element with 1/4 the resistance of the original. It will draw 4x the current giving 4x the heat output. 4 * 385 = 1540W which is more useful.

That's a great idea! There are several points I could bridge in the wire to shorten the circuit. The middle will go dead but the edges will heat up. I think I won't cut quite so much as you suggest out because the variac has a minimum load of 10Ohms. Still I think that basic concept should work.

Quote: Originally posted by densest

The points of connection between nichrome and copper are where it will be most likely to burn out. The copper will oxidize and the nichrome will get especially hot because there's nothing draining heat away from that point. Doubling the nichrome there would have been a good idea. i.e. wind an extra strand over the joint and a couple of cm into the heating section. McMaster-Carr (and other suppliers) sell nickel wire covered in glass fiber and nickel clamp connectors for this sort of application. Join the copper to the nickel at least 10cm away from the hot area.

I'll keep that in mind, but there's not any copper in this at all. The nichrome is connected to a ceramic terminal connector with I'm guessing nickel or steel terminals. I'll check the resistance while its hot and see how much the temp changes it.

@peach - I thought about using a step up transformer to increase the voltage but the variac is only rated for 120 V, so that won't work. When I make my next heater I will have all these lessons under my belt and make something really good!

Something else I was wondering was since I wanted it to be explosion proof, whether the exposed heating element could trigger an explosion, since it gets quite hot. Well above the flash point of some of the more flammable things I may be heating. This is why I buried it under a layer of cement, but it pokes up where it connects to the terminals. Is this a serious risk or am I being paranoid?

peach - 5-7-2010 at 16:07

[edit: I'm glad to see you're using a transformer; providing an isolated floating voltage. I expect someone may be tempted to try it directly from the mains. That will end in pain]

You can still use a step up transformer after the variac, because it'll only see 120V on it's own coil. Just be sure to put it AFTER the variac.

The flash point of things gets confused with another term, the auto-ignition temperature.

The first is the temperature at which the chemical has a vapor pressure high enough that'll ignite if subjected to a source of ignition in the air.

Auto-ignition is when something spontaneously ignites without a spark, and it's often at a much higher temperature.

For your plate, you'd want to be more worried about sparks coming from the contacts on the variac as you dial the voltages around. Or from the light switch.

But having white hot wires poking through the cement might be an issue too. I'm also wondering how long those bits will last with the temperature difference, alkaline nature of the cement and all that jazz.

Check the auto-ignition temperatures of the things you're dealing with prior to heating them up.

Remember, industry only uses explosion proof plates in case there's an accident, not as a form of protection. The first thing they do is try to stop that flammable / explosive thing getting into the atmosphere in the first place. This is analogous to my ideas on fume hoods versus wash bottles. One stops a problem from becoming a problem (wash head), the other attempts to fix it once it is a problem (fume hood).

Now you've given me a burning desire to make some fireworks. Maybe I could use one to distract this damn moth that keeps harassing me.

[Edited on 6-7-2010 by peach]

densest - 6-7-2010 at 08:55

@crucible - it's easy to lower the resistance without making any part go dead.

Right now you have mains1 --/\/\/\/\/\/\/-- mains2

If you tap into the middle and join the ends so end1 & end2 go to mains1 & tap goes to mains2,

end1--/\/\/\/-tap-\/\/\/\---end2

then the resistance is 1/4 the original.

You can tap in 2 places and get 1/9 the resistance by joining end1 & tap2, end2 & tap1.

Good luck!

[Edited on 6-7-2010 by densest]

cnidocyte - 7-7-2010 at 13:46

This is honestly one of the coolest improvs I've ever seen. Electronics is one of my many hobbies so I may build one of these myself some day.

Crucible - 10-7-2010 at 20:31

@densest - Ah, I get you now. But that would require a more thorough rewiring than a simple bridge and that's going to be hard to do with the cement already set.

@peach - I hadn't thought of that! Yeah, I could step up after the variac. I'd need the right kind of transformer though, but I could easily double the temps by doubling voltage. Since both 110 and 220 are very common voltages, there must be a simple transformer I could use.

That could end up being the best of both worlds, because one of the things I was concerned with about having a super high temp hotplate was losing granularity of control. But I can take a transformer out of circuit for low temp stuff and only use it when I need more heat. Do you have any suggestions for what kind of transformer I would want specifically?

Yeah, auto-ignition is what I was thinking of. So far I've only used it to distill acetone, and I sealed up the joints but I could still smell it a little. At some point I plan to fraction some pentane which is quite a lot more flammable. Seeing those wires glowing red hot made me a little concerned. I've already eliminated pretty much every other source of ignition. I turned the breakers for the room off and run an extension cord from outside to the plate. And the variac itself is solid state and a good 6 feet away from anything combustible. The water pump is, well, under water. So my biggest worry is those wires.

At some point I want to slide a brushless fan underneath it with rare earth magnets glued to the blades so I can use a magnetic stir bar. I'm just not sure if the fan I had in mind is going to take the heat coming off the bottom of the hot plate, and if its too far down, the magnets lose strength to "reach" the top of the plate.

Crucible - 10-7-2010 at 20:38

Quote: Originally posted by cnidocyte

This is honestly one of the coolest improvs I've ever seen. Electronics is one of my many hobbies so I may build one of these myself some day.

Thanks! I put a lot of work into it. The weakest link temp-wise is the ceramic terminal connector which is still rated at 800C, so once I get the electrical stuff worked out I should be able to get it quite hot. It has definitely been a learning experience though, all the materials are quite "finicky" to work with and take a special manner to get them to take on the shape I had in mind. A lot of mistakes were made and redone! I'm going to make a heating mantle next, using the same type of materials. If you are looking for something a little easier to make, I'd suggest taking a cooking hotplate and take the temp control knob out and rewire the heating element to a variac.

Contrabasso - 11-7-2010 at 00:16

When I had a hotplate stirrer, the element was a ceramic ring under an alloy plate, and the spinning magnet was in the centre of that toroid.

You could increase the power and temperature of your hotplate by adding a 120 - 180 transformer if you can locate one. but that would also change the life expectancy too. Having the connector block on the top seems strange as that's where you want to put the beaker or flask. Also it means that you have mains live contacts easily within your touch range when operation the hotplate, Again better if the contacts were under the hearing surface.

Distilling volatile and flammable organics in any quantity is better done very carefully and possible with steam heating -even raise steam in another room and pipe it to where you need it. I've seen one lab fire through a vapour flash over, one is quite enough.

JohnWW - 11-7-2010 at 00:31

Has anyone bothered to look up the patent literature on hotplates?

zed - 13-7-2010 at 12:12

Umm. I've generally found that standard heating elements, similar to those from an electric stove top, work pretty well. I used to just take a portable hot plate unit, and remove the temp control, replacing it with a soldered (hard wired) connection.

Thereafter, I would control the temperature remotely. Don't remember what kind of voltage controller I was using, but it wasn't anything as fancy as a variac. Worked OK I guess. I'm still alive, and I'm not hard to look at.

densest - 14-7-2010 at 07:41

I'll vote with zed - $10 to $15 at several local "remaindered general merchandise" stores. Spilling aqua regia on one usually means replacing it, though, since the cheap ones have metal covers over the resistance wire. Next time I'll look for a ceramic top one.

I recently got a very inexpensive stirring hot plate on EBay. They're around if you scrounge.

zed - 14-7-2010 at 18:31

I actually got TWO very fancy name-brand stirring hot-plates for under a hundred bucks total....including shipping. After a few minor adjustments, both worked perfectly.

Now, neither one is explosion proof. For explosion proof, you pay a big premium.

For a hundred bucks or so, I could probably build an explosion proof unit.... inside the shell of a standard stir-plate unit.. Probably use a Gast 2AM airmotor for the stirring component. They usually run at 1000 to 3000 RPM and they don't require a huge flow of compressed air to operate. 1000 RPM is the "sweet spot" for entraining gas.

Once again, I'd remove the temp controller, and control the heat remotely. Then, I'd use a special grounding wire for the whole unit . Static build-up being a possibility. A spark from any conceivable source, is still a spark.

Naturally, using such an explosion proof unit requires compressed air. So..... you need a pretty good compressor to produce it. Another expense. Still, you need that compressor anyway.....to power your explosion-proof overhead stirrer.

peach - 14-7-2010 at 19:24

I keep seeing small mountains of IKA plates turn up on eGay that look like someone has set about them with a hammer. Oddly, it always seems to be the left hand dial that's snapped off.

Saying that, my plate now looks like Frankenstein's monster, but still functions fine (and I suspect those IKAs either work or can be coaxed into doing so). In fact, my filthy plate makes less noise than it did when I got it (the magnet used to rub against the stand offs). I wish more lab gear was quiet. Once I have the the recirculating cooler going and the hotplate, it's a constant buzzing for hours on end. The plate in particular BuzzzzzzzZzes loudly when it's turned down low, with the PWM circuit chopping away.

The cooker hotplate idea is a good one. I used to use the kitchen cooker when I ran out of plate space.

This obviously isn't a guarantee against blowing yourself up, but computer fans are DC brushless motors. From the wiki on electric motors;

Quote:

Brushless motors have no chance of sparking, unlike brushed motors, making them better suited to environments with volatile chemicals and fuels.

Hotplates tend to use ceramic / ferrite magnets, because the Curie temperature (the point at which they loose their magnetic appeal) is around 460C. They're also not very strong.

You could use one of the high grades of neodymium magnets and simply place the fan further away from the heat source. If you're distilling solvents and such, heat isn't going to be that much of a problem. My stir bar will sometimes jam because the motor can't turn against the friction, not because the magnet can't hold the coupling force. And it's a good thing it does a lot of the time, because the alternative is that the flask starts trying to spin.

Another option for a brushless design would be to directly couple to the magnet it's self and use that as your rotor. Pancake motors found in disc drives and things like that do this. They have coils that produce the rotating field.

There are also 'mag drive' pumps, which are used a lot in the chemical industry, which use magnets in the rotor it's self and then produce the field around the impeller. That way, the impeller can be made of something like PTFE and the pump head doesn't need dynamic shaft seals.

Even garden water features use mag drive impellers.

There are tons of guys using computer fans to stir things now, varying from just plugging them into an old computer power supply or a wall adapter to building little LM317 circuits to control them. A lot of them are brewing alcohol.

There's also a PDF floating around that someone on DIY Audio wrote about taking apart disc drive motors and using them to power turntables; as the rotation has to be very smooth and at a constant rate to read the data correctly. Disc drive motors, as a result, have a lot of circuitry already attached to control the coils. I think they're actually closed loop regulated (which means the circuit can tell how fast the element is actually spinning, unlike most hotplate / stirrers). They have some crazy number of phases for the coils.

Cheapest PC fan on ePimp-wegotsyabankdetails.co.uk is £1.60, delivered. That's what IKA charge per sentence they answer a commoner's email with.

If you were running a small air powered motor, you might be able to drive it from a fridge compressor. Which is a lot quieter than a normal piston. Again, the piston is powered by a brushless method. They'll easily produce 100psi and I've actually heard guys talking about hundreds of PSI from them. But, admittedly, I've never really looked at the CFM for air motors. Other than when I looked at air tools for the workshop and discovered they need an 8HP compressor (or something similar) to produce less power than a £5 budget line, no name, corded drill. That, combined with high pressure spraying, wastes a monumental amount of energy industrially.

Here's the DIY Audio thread where they're rabbiting on about floppies

And this is an example of a fan stir plate from Mad Zymurgists (he has a photo guide on this link);

[Edited on 15-7-2010 by peach]

zed - 14-7-2010 at 20:18

Neodymium Magnets really can't take much heat. To my recollection they fail at 80 C or so. Well below the BP of water. Better to use massive magnets inside your stir plate... Massive magnets that have a lot of heat resistance.

Samarium based magnets are pretty strong, but they are also pretty expensive. No can do.

Likewise, Dyna-mags and other sealed overhead magnetic-drive stirring units are prohibitively expensive.

Parr and their ilk, want thousands of dollars for a sealed magnetically coupled stirring unit.

The guts of these units, the neodymium magnets? Well, they're cheap as dirt.

Design an easy to build, cooled, sealed stirring unit, utilizing neodymium magnets.
It doesn't have to be incredibly durable. Just needs to be able to withstand high internal pressures, and corrosive chemicals. A unit that would run a hundred hours or so before needing a rebuild, would be fine.

You will become a hero.

[Edited on 15-7-2010 by zed]

JohnWW - 15-7-2010 at 04:51

If Nd magnets can only stand 80ºC or so, the temperatures inside many hard drives that use them may come perilously close to causing their failure; unless they are of a different composition from those used in modern magnetic stirrers, the hard-drive ones being Nd2Fe14B according to another thread, http://www.sciencemadness.org/talk/viewthread.php?tid=14145 , with some of the Nd replaced by Pr. They appear to owe their strong magnetic moment to the two surplus 4s and one surplus 3d electrons from the Fe atoms (left with five unpaired 3d electrons) entering the vacant 4f orbitals in Nd (which has three unpaired 4f electrons of its own, with four of the seven 4f orbitals vacant) (or Pr, with two unpaired 4f electrons of its own and five vacant 4f orbitals), so as to maximize the number of unpaired electrons.

[Edited on 16-7-10 by JohnWW]

zed - 15-7-2010 at 10:31

Thermal Characteristics

Nd Material Type-------Maximum Operating Temp

N---------0.12----------------80ºC (176ºF)
NM-------0.12----------------100ºC (212ºF)
NH-------0.11----------------120ºC (248ºF)
NSH-----0.10----------------150ºC (302ºF)
NUH-----0.10----------------180ºC (356ºF)
NEH-----0.10----------------200ºC (392ºF)
NZ-------0.10----------------200ºC (392ºF)

The strongest magnetic materials are imbued with the least heat resistance. Too bad! The cheap, super-strong, Nd type magnets that are commonly offered for sale are N-type.

Dyna-mag type units, must be equipt with a cooling water jacket, to keep their Nd magnets from frying.

Also of note, Nd magnets are very adversely effected by Hydrogen. Simple plating is not effective protection. Gotta be hermetically sealed, or otherwise isolated from H2. They don't blow up or anything. They just moulder away.

[Edited on 15-7-2010 by zed]

Contrabasso - 15-7-2010 at 11:36

Heaters for a stirrer motor NEED to be torroidal so that the magnets can get close to the stirrer bar. with your disk shaped heater the rotating magnets will be 20 - 30mm below the stirrer bar and the coupling will always be weak and delivered power may be too low to stir effectively.

zed - 15-7-2010 at 14:05

In stir plates, the strength of magnetic coupling is always a problem. Really vigorous stirring requires a very strong coupling. Even when using an Nd "supermagnet" as a stir-bar, tweeking may be required. At a thousand RPM or so, the stir-bar tends to uncouple and break loose.

Solution? Closer proximity, bigger magnets, and maybe a bigger stir bar.

Torroidal heating sounds good. But, not too much heating.... Sigh! My Nd stir bar fails at high temperatures.

If very much heat, pressure, viscosity, or high RPM is involved.....Ya gotta look elsewhere. Overhead stirring, or sealed overhead stirring, in a sealed reactor.

http://www.pressureproductsindustries.com/mixers/dynamag.htm...

http://www.parrinst.com/default.cfm?Page_ID=164

Crucible, I like the cement/nichrome hot plate quite a lot. It has a rustic, sort of pre-Columbian look. It might be difficult to make it perform as a usable lab-type hot-plate, but it has real potential as kitchen "ART". It would make a damn fine coffee pot warmer, as is. Might be able to sell millions of units, all over the world. Jump on it, before Martha Stewart steals your idea.

[Edited on 15-7-2010 by zed]

[Edited on 15-7-2010 by zed]

Contrabasso - 16-7-2010 at 00:56

At my old work lab, we used always to be short of stirrer magnets so we used to put a bit of steel wire in a glass tube, heat and crimp the ends. You soon learned to bring a stirrer up to speed gently! or the bar flew off

peach - 16-7-2010 at 13:00

Quote: Originally posted by zed

In stir plates, the strength of magnetic coupling is always a problem. Really vigorous stirring requires a very strong coupling. Even when using an Nd "supermagnet" as a stir-bar, tweeking may be required. At a thousand RPM or so, the stir-bar tends to uncouple and break loose.

That's turbulence in the fluid. The same thing happens in fusion reactors. The goal is to keep the plasma flowing in a perfectly smooth circle, with the ions spiralling around the field lines of the toroid. Real world turbulence ruins that and they fly off the field lines.

I also realized my point about stronger magnets further away is kind of dumb. The field strength drops by the ?cubic? law, so the magnets would need to be ?four times? stronger per unit distance?

[Edited on 16-7-2010 by peach]

zed - 16-7-2010 at 15:58

The drop off in magnetic field strength/vs distance is shocking. I have a stainless steel vessel that has a 3/4 inch thick bottom. I would like to use it with magnetic stirring. No way.

The stir bar, ends up being too far away from my stir-plate's drive magnets, for strong coupling.

No problem with thin-walled stainless steel vessels. The magnetic field ignores the stainless steel. So, magnetic stirring works fine.

It's the distance, not the material. Three quarters of an inch, might not seem like much, but in this case it's an insurmountable obstacle to success.

JohnWW - 16-7-2010 at 17:18

Although the attractive force due to electrostatic or magnetic monopoles, like gravity, falls off with the square of distance 1/r², with spherical equipotential surfaces, with oppositely-charged mutually-attracting dipoles in close proximity (which is usually the case) the attractive force falls off with the cube of distance, 1/r³, from a fair distance from the dipole, with equipotential surfaces planar cross-sections of which are "ovals of Cassini", 4th-degree curves. This is dealt with in 200-level Applied Mathematics and Engineering Mathematics.

Crucible - 16-7-2010 at 19:04

Quote: Originally posted by zed

Hmmm, I'm not sure my PC case fan I'll be using as the stir motor gets up to 1000 RPM, I'm guessing it will top out at 350 RPM before I even mount magnets on it (extra weight).

Based on the comments, I decided to do a quick test. I have 2 types of magnets to work with. Large heavy hard drive magnets and small light rare earth magnets. The thickness of the hot plate is about 1.5" and I'm guessing I'll need strong magnetic coupling at a minimum of 2" to account for an airgap and the thickness of the glassware.

When I position either type of magnet under a ferrous object (a screw in this case) I can get it to move (ie magnetic force > friction) from a distance of no more than 1.5" which is too little. However, if I position the small magnet on top and the large one on bottom the effective range is doubled and I get very strong coupling at 3" or more, which should be plenty. Assuming my stir bar is as strong of a magnet as my small rare earth, which it probably won't be.

So this plan might actually work if I can find a suitably strong magnetic stir bar, or find a way to seal my little magnets in PTFE and make them into a stir bar.

Quote: Originally posted by zed

Torroidal heating sounds good. But, not too much heating.... Sigh! My Nd stir bar fails at high temperatures.

Yeah, I'd forgotten heat is the enemy of magnets. Thanks for the reminder. I think I will stick to trying the magnetic stirring for 60C or less.

Quote: Originally posted by zed

If very much heat, pressure, viscosity, or high RPM is involved.....Ya gotta look elsewhere. Overhead stirring, or sealed overhead stirring, in a sealed reactor.

If I can't get magnetic stirring to work or need high heat, I'll need to invest in a 2 or 3 neck RB and an overhead stirrer. Though I may experiment with aluminum or mylar heat shielding around the magnets to see if that works.

Quote: Originally posted by zed

Crucible, I like the cement/nichrome hot plate quite a lot. It has a rustic, sort of pre-Columbian look. It might be difficult to make it perform as a usable lab-type hot-plate, but it has real potential as kitchen "ART". It would make a damn fine coffee pot warmer, as is. Might be able to sell millions of units, all over the world. Jump on it, before Martha Stewart steals your idea.

LOL! Thanks zed. pre-Columbian huh? "Deep in the ruins of Maccu Pichu archaeologists have discovered an ancient heater which actually runs on electricity, right next to some primitive glassware which surprisingly has standard 24/40 joints" Hahaha

I'll rush them into mass production just as soon as you tell me where to find people willing to spend $150-$250 on one. The cement alone cost me $50, the wire and accessories another $35 and I don't even want to think about how much I paid for that solid state variac! But I still love her anyway, and if I ever get the design perfected I may just sell them, to chem enthusiasts obviously. Not much potential in a $150 coffee heater!

And then I can start work on the perfect design for an enclosed overhead stirrer.

densest - 17-7-2010 at 10:26

The market for "concept" = insanely overpriced "decorative" = ugly house things is recovering. The too-rich are also too-dumb... The cement is probably much cheaper in bulk or as a mix. I know that "furnace cement" can be expensive at a hardware store and much cheaper as "kiln cement" or "castable refractory" in bulk. Solid state variac = dimmer 600W @ $15. Nichrome wire at 28ga (4.25 ohms/foot) is about $0.17/foot.

I've successfully fused magnets, etc., into PTFE tubing. The keys are temperature control and speed.

PTFE doesn't really melt well. It does become very soft and jellylike just below beginning to decompose. At that temperature it can be molded and fused to itself. The window between too cold to fuse and decomposing is narrow.

Procedure: cut a length of PTFE tubing 3 cm longer than the object to be encapsulated. Get a pair of pliers with jaws wider than 2x the tubing diameter ready where you can pick them up very quickly. Put the object in the tubing and center it. Using hot air (either a hot air gun or over -not- in a flame or hot plate) gently warm one end over a 1-1.5 cm length. Turn the tube so it soaks in heat all around the diameter. It will lose opacity to become translucent to almost transparent. At that time grab the pliers and compress the end of the hot area to seal it. Don't apply significant force - the plastic is very soft. Leave a short length of hot plastic between the seal and the cold part to take up stress so that the result tapers between the round area and the flattened end. Cool, then do the other end.

One could make a conical/hemispherical tool or a pair of half-cones or half-hemispheres on tweezers to make a rounded end which would probably be stronger.

The procedure takes less than 5 minutes from start to finish.

[Edited on 17-7-2010 by densest]

peach - 17-7-2010 at 15:35

I don't wish to be a negative nancy, but have you tested those ideas everyday with numerous different chemistries? Like the seals on the ends of the tubing via various heats and chemical agents? Does it stir properly? I can stain a genuine PTFE bar black without too much effort. And I've beaten the shit out of my stir plate despite being careful with it. It looks like a piece of modern art.

Remember, crucible is talking about explosion proof gear. As I've said in another thread, I have enough problems of my own using genuine PTFE molded bars or genuine plates; and I'm not dealing with explosion hazards. Trying to improvise explosion proof gear, doing acetone as your first distillation and having trouble getting it to boil are concerning given the end goal. Someone doing that should easily be able to accomplish the former. That plate doesn't seem to have been over 100C yet, the boiling point of the water that's inevitably trapped within the cement. Any potter or caster will tell you, trapped moisture is an issue. When you have a litre or two of something highly flammable or explosive balanced on top of it, it's serious issue; e.g. the cement burst, smashing the flask and exposing the contents to the red hot element.

If you've ever light a fire on concrete, you may know what I'm getting at. I've done this with concrete decades upon decades old and had it pop like there were fireworks in it as the trapped moisture boiled.

I am genuinely proud of people DIY'ing gear themselves, it really is in the spirit of things. But I worry that crucible will assume it can perform in the same environments all the certified gear can. There are so many variables to think of to jump from distilling acetone to explosions, "will that seal break", "will my terminals get too hot", "will the cement burst", "will the variac spark", the list goes on for so long. To design that kind of thing, you need to understand the fundamentals of what makes something explode. And knowing the difference between a flash point and autoignition temperature is one of the first on that list.

Before I came to this forum, I spent about seven or eight years up each night reading high end electronics forums for six or nine hours a night. I would still hesitate working with explosion hazards. You have to have everything nailed to the floor in terms of the variables.

Please take some photos of your PTFE encapsulated things. Not so's I can complain and pick holes. I've heat sealed things before myself and I'm genuinely interested to see how it works with PTFE. Watch those HF fumes.

I'd get rid of the feet on this design. If the cement ever fails, the feet will cause it to cave in on it's self, potentially breaking the flask. If it were supported by a continuous surface, that wouldn't happen.

[Edited on 18-7-2010 by peach]

densest - 17-7-2010 at 21:30

@peach - you're asking about risks and risk evaluation. Edit: there's a "Bonding PTFE" thread on this site with good info.

I'll post pictures as soon as the *#@#$ box with the stir bars surfaces - no promises about when :mad:

because it's well & truly buried. The lab where I used it had to be shut down due to the landlord doing bizarre things unrelated to the lab. We, the tenants, left very quickly when we heard about the problems, not wanting to be caught in legal crossfire.

Anyway - risks: I've done 3 stir bars the way I described. I examined all of them under magnification for cracks, wrinkles, and other evidence of strains. The fused length was 1cm. I tested them by heating them in cold water to boiling while watching for bubbles, then cooling to room temperature. If there were leaks, either air would bubble out or water would flow in. Neither happened. So, my judgment is that for temperatures 0C-100C, they're OK. One risk is acid reacting with metal inside. So in alkaline, anhydrous, neutral, or mildly acid environments, the risk is low. Strongly oxidizing (especially chlorine) acidic environments are the biggest risk for metal reactions. Above about 100C, any water inside would vaporize risking bursting.

PTFE is very inert. It is vulnerable to mechanical damage, very strong bases, and molten alkali metals. Dunno about molten Ca but I wouldn't risk it. My homemade stir bars are weaker mechanically than commercially made ones, or so I must assume, so I won't use them if there's any risk of mechanical damage through abrasion, impact, etc., or heating above about 95C, or in strong bases. I inspect them for damage after each use.

"Explosion proof" is about not providing activation energy for flammable mixtures and not propagating flames or sparks. It doesn't mean "invulnerable to any damage" - mechanical & other chemical risks are different problems.

I think discussion about risks is very important - as long as risks are at least relatively quantified and carefully analyzed.

[Edited on 18-7-2010 by densest]

Crucible - 19-7-2010 at 17:58

@peach - I appreciate the concern, but that is the reason I posted this thread, to work out all the kinks and get input from others whose knowledge exceeds my own. I think I have things pretty close to nailed down at this point. Unless there are some other variables I haven't considered. But lets go down the list of things I've already taken into consideration.

Water in the cement - there was a good bit of moisture in the cement when it was first made. I always test my gear before using it "live". I spent several hours calibrating the variac and testing the plate. Before I ever used it to heat anything, I heated it to 50C for 3 hours continuous, let it cool, then heat it again to 90C for 3 hours continuous. There is a peculiar smell that moisture seeping out of cement carries, and I got an ounce or two of water out of it before it "ran dry". Since then I used it to distill the acetone (also an explosion hazard) where it ran at 60+ C for greater than 6 hours continuous. I think if there is any water left it would be very, very little. And since I won't actually need to push it to 100C to boil flammable solvents, there is no reason to think that any residual water would cause a catastrophic blow out. Even if there was, it is pretty unlikely it would break the flask, because I'm using a double boiler setup and all the glassware is supported on stands NOT by the hotplate. So if the hot plate cracks, yes there will be exposed heating element but that's it. And as for structural integrity, I've done some informal stress tests with the cement and it takes quite a bit of force to break it. The likelihood of it breaking under its own weight or the weight of the water pot is quite slim IMO. The top is a full 1.5" thick and it took >10 lbs of force just to break a .5" piece.

As for the exposed heating element, I'm planning to dump some more cement on it just to cover it up. However, just on the chance that there might be something left exposed or the cement may crack and expose some during operation, I'm also planning to use a fan pointed at the hot plate to blow any fumes away. I'll have any potentially spark producing objects upwind and the great outdoors downwind. I can't conceive of any possible scenario in which enough solvent and/or fumes could come into contact with anything capable of igniting them under these circumstances (remember I will have turned off the breakers for the entire room also). And I don't think that because my equipment is DIY that it is unsuitable or inferior. If anything I know what it is capable of and any potential hazards better than I would some second hand piece of manufactured labware which might have been subjected to all sorts of dangerous conditions I'm not aware of, mechanical shocks that may have ruptured the internal seals, chemicals that may have corroded the insulation on the wires.

But if you think there is any other thing I've overlooked, I'm all ears.

peach - 20-7-2010 at 10:48

@Denset

Sorry buddy, I was in a bad mood when I posted that and was complaining more than need be.

The reason I'm skeptical is that I know one of the major problems with PTFE is that it can't be thermoformed. Indicating that the bonds you've produced are mechanical, as opposed to chemical.

I haven't read that thread, so I can't speak with much certainty, but I suspect industry would be using this method if it produced a true PTFE seal. Instead, they've spent a huge amount of time and money, and induced strong risks, etching the surfaces and then gluing them together with less than ideal seals.

If you want to do that kind of thing, look at PFA. That's basically Teflon, but thermoformable. I think FEP is as well. FEP is PTFE with a carbon sticking out the branch, but it's fully fluorinated. PFA is the same but the carbon is linked to the chain with an oxygen. They're used as liners for things like Tygon tubing when it needs very high chemical resistance. PTFE tubing is too rigid to use for similar applications a lot of the time and can only be used as a pushed through liner membrane for tubing.

I'd use PTFE for corrosive gas handling, but it's fairly rigid stuff (so it'll be a pain to setup and try to topple the glassware) and it's opaque.

I don't think your stir bar needs a super strong magnet in it. Mine isn't all that strong and I can stall the motor in a commercial plate with thick slurries. That's were overhead stirring comes in. And that's where you have to start messing around a lot more. It's easier and cheaper to use a bigger flask and throw some more solvent in most of the time.

You are correct about risk concern where it's needed. If you see some of my other posts, you'll see I'm quite the opposite of a scaried cat and have been suggesting the toxic label doesn't apply to things most other people consider toxic.

It's just, as I hint at, explosion proof gear usually means someone is doing something incredibly dangerous. Where a mistake will torch an entire facility. Without be being clear what crucible is planning to do or what experience he has with it, it's hard to downgrade that on the basis of "I know this guy, I know he knows and is asking to be sure, I know he can probably do it without asking". Also, I don't think crucible genuinely needs explosion proof gear. Distilling acetone is something countless labs do with nothing close to explosion proof hotplates.

@Crucible
It's good to know you're using a bath, that'll solve the fire / explosion risk if the plate cracks.

But you are kind of not being very consistent, as you were requesting a thousand degrees on the plate and have now swapped to 100, an order of magnitude less.

There are numerous solvents that are theoretically flammable, but next to impossible to actually ignite in the real world. Like acetone. It's very easy to distill acetone with a normal plate and not have any problems.

If the solvent is building up in the atmosphere, you're wasting solvent, producing a health hazard and risking a fire or explosion. It means your condenser isn't cool enough and / or doesn't have a long enough surface area.

Acetone is miscible in water. So, venting your receiver bend through a wash composed of water would trap the vapours before they got airborne.

Explosion proof gear is usually used in industry were an accidental escape would result in the entire warehouse instantly going up. The goal is to stop it getting airborne in the first place. It's the old prevention over cure theory.

Don't rely on things to prevent problems once they arise (like fume hoods and explosion proof gear). Get rid of the problem before it becomes one.

A good way to stay safe is to look at the concentrations of the chemical that can be readily smelt, then at the concentration needed to create a fire or explosion. That way you'll know "I can smell it, so it can ignite" or that it's getting close to a problem.

Flammable chemicals have a range of dilutions they'll ignite in, and others that won't. You can often need quite a lot of the fuel present in the atmosphere for it to start burning, far more than is needed to smell it. And, once it gets concentrated, it often won't burn anymore. You can fill a room with highly flammable chemicals and, provided they've displaced a good percentage of the oxygen, you could throw a light blowtorch in there and nothing would happen.

The fire triangle is a good thing to remember. But add to that, the fuel, oxidizer and ignition source have specific parameters set on them as well; e.g. the difference between flash points and autoignition. It's not a static thing. With chemicals and odd environments like chemists experience, you can't simply tick them off. You have to think about each and how it varies, because sometimes they're ticked on or off when they wouldn't be in the normal world. It's quite dynamic.

[Edited on 21-7-2010 by peach]

densest - 20-7-2010 at 22:34

@peach - Thanks, yes, there are bad days and bad weeks... sometimes bad years.

Agreed, yes, the bonds are mechanical. I'm only asking that the joint be water/solvent tight for very low pressures without other mechanical stresses. Making my own NaNapthalide is a project for some other day. I wouldn't make a beaker the way I suggested, no, it would be a disaster!

PFA is nice... I use PFA fittings to attach my PTFE tubing to various pieces of equipment. They're only mildly exorbitant from McMaster-Carr. Yes, I use the PTFE tubing for corrosive gases like Cl2, HCl or HNO3/NO2 & friends. I got some thin-wall tubing so it's not as stiff or opaque as it might be. It's often necessary to use a couple of extra feet of tubing and loose coils of it to get around the stiffness. And lots of extra horizontal & vertical rods & clamps to make sure I don't get another bath in aqua regia. Two were enough.

I've found that the motors used in stirrers/stirring hot plates are very badly matched in torque and speed to the requirements of the job. The motors are used at speeds well below their optimum. One wants 60-600 RPM or so, and the motors are 1800/3600 RPM AC motors slowed down by SCR controllers. To fix this would require either a multipole motor (expensive), gearing (expensive), or a DC motor and control (also expensive).

One can get surplus DC gearmotors for $5 - $25 with speeds in the appropriate range(s) very easily. For a DIY overhead stirrer they work very well. Getting a brushless one is a little harder if one needs explosion-proof... and if I needed explosion proof equipment, I should very carefully reexamine what I was doing!

peach - 21-7-2010 at 00:09

I found your perfect chrimbo present Densest....

>>>>>>>>>>>>>Christmas for Densest<<<<<<<<<<<<

I spent a few hours hunting for some FEP or FPA tubing to fulfil my corrosive gas needs and found some really cheap (less than $700 a roll), but I don't think it's a wise choice for glassware. I contacted one seller who confirmed this for me (that it'd kink all the time and not bend very easily). I also did some significant editing of the wiki's on PTFE, FEP, PFA and ETFE, as they didn't really explain which was better for what.

You're right about stirrer motors as well. By the time I have mine spinning at full speed, the bar will frequently detach it's self due to turbulence. A big stir bar rocketing off it's axis inside a flask full of something corrosive, nasty, capable of staining the ass out of something or pretty much anything other than water is not a fun thing. Not counting the cost of replacing any glass it breaks.

My no name plate was about £150 - £175 I think. When it arrived, the magnet was skuffing against the baffles it went through, making a 'chif, chif, chif' noise. That went away.

But now something seems to be going wrong with the stir circuit. When I set it to full, I can hear something that sounds a lot like arcing.

I'm a little concerned about it actually. I'm not an electronics god, but I know that when I turn something off, the neon lights connected to that switch shouldn't keep glowing (like they're leaking to earth) and the PWM circuit was making a questionable amount of noise before the arcing sound appeared.

Saying that, it's worked fine up until now and still does, for now. It will go hotter than an IKA (which is handy for drying salts out) and cost a whole lot less.

[Edited on 21-7-2010 by peach]

densest - 21-7-2010 at 13:27

Oooooh! Must have

Thanks!

The same person is selling 1500W stovetop heating elements for $15(?) or so. Dunno how thick they are or whether the top plate is magnetic... still, it's about the right size & power.

I have another (non-stirring) hot plate which is infrared/radiant heat. Not suitable for metallic or reflective containers. Works very well with glass. It's a non-contact approach.

I'll keep my eyes out for FEP or PFA. As I said, the PTFE tubing (0.031" wall) is quite usable if you're willing to observe its minimum bend radius. It was very inexpensive.

I saw huge quantities of fluorocarbon pipe go for a song about 5 years ago :mad:

I had no place to put it.

[Edited on 21-7-2010 by densest]

quicksilver - 21-7-2010 at 15:54

A close friend had access to a damn nice lab and had some very expensive, quality equipment. One thing I noticed was that really quality hot-plates were not integrated with stirring mechanisms: they were stand-alone items. Another issue was that the heating surface was QUITE THICK (& there was a was a label upon the hot-plate that directed the user to wait before using so as to find stability at proper temp). The electronics of the hot-plate were also not cheap & very high wattage. Checking Mouser [Electronics] for the cost of the rheostat found it to be enormously expensive for that single item. Other parts were uniquely high-end. No cost cutting venues were in place what so ever; it was impressive. Which leads me to believe that one can have standard quality & decent utility or one can have first rate quality but need to pay the price.
I've own several hot-plate-stirring devices and I really don't think they ever were so well made that I wasn't forced to check (& re-check) the temp constantly as changes of as much as 10 degrees were not uncommon. I've owned YellowLine & Corning plates but the better, high-end Barnstead plates can be found to also have schematics available.

Crucible - 23-7-2010 at 18:09

@peach I don't know that I'll ever get my plate up to 1000C, but I designed it so that I would have the option. I'm a big fan of having few pieces of equipment that have a wide range of capabilities. Sorry if that confused you.

You might be right that I don't need a super strong magnet, but I'll need a regular stir bar to test the magnetic coupling and have the option to seal some strong magnets into PTFE if I need to. I'm all for using the option that works with least cost/effort.

My main concern with explosions was indeed volatile solvents (ether, petrol distillates, etc). And my concern with that is joints leaking. Having some fumes leak, ignite, and flash back shattering all my precious glassware, spread flaming solvent everywhere, and shower me with broken glass would really ruin my day. I always try to prepare for the worst case scenario, so that's why I went for explosion proof. Isn't that what it's for?

peach - 24-7-2010 at 12:13

The following will read as nitpicking and bitching. It is not intended as such, I'm curious about the details.

Quote:

One thing I noticed was that really quality hot-plates were not integrated with stirring mechanisms: they were stand-alone items.

I've seen this as well. I think it's that they expect a lot of their buyers will be making use of overhead stirrers. It could also be laziness. I've cracked open a plate that cost $1k a piece. It wasn't anything that exciting inside.

Quote:

Another issue was that the heating surface was QUITE THICK (& there was a was a label upon the hot-plate that directed the user to wait before using so as to find stability at proper temp).

That again sounds like they don't have faith in their own product. Do you mean wait for it to heat up and then put the glass on it? That's asking for it to pop. The thickness of the plate would only add stability to the temperature if they haven't added it electronically.

Quote:

The electronics of the hot-plate were also not cheap & very high wattage.

I'm not trying to be rude, but how long ago did you observe the insides of these? I can now buy a transistor that'll dump the entire load of my house through it's self for not very much money at all. My own cheap plate is a higher wattage than the $1k one I bought.

We just got a fancy induction cooker top that's got a fan cooled PWM supply in it that'll regulate something like 6kW (ten times more than a normal lab plate) and it's full of alarms for when the surface is hot, timers and locks to stop kids burning themselves. That was about £200 I think. The miracle of solid-state.

Quote:

Checking Mouser [Electronics] for the cost of the rheostat found it to be enormously expensive for that single item

This is where I worry and what made me question the age of the plates, as a rheostat is (to my knowledge) and massively outdated method of controlling a load like that. There are linear regulators and PWM now. I can put together something that'll control similar loads with a handful of components that cost a few dollars. A rheostat also wastes a ton of energy when small powers are needed at the load, and inherently wastes power whatever load it's set at as it dumps some of the current through it's self.

A resistive divider was literally the first circuit we were introduced to in secondary school, when I was 14-16.

A rheostat also means no active, closed loop feedback and therefore, no temperature regulation. Which goes back to the point about plate thickness and waiting for it to warm up.

Quote:

No cost cutting venues were in place what so ever; it was impressive. Which leads me to believe that one can have standard quality & decent utility or one can have first rate quality but need to pay the price.

I'm saying this entirely willing for you to tell me it's totally wrong, but that sounds a lot like a very lazy method of designing things. And then passing the price for the high cost components onto the customers. A lot of companies still do this now. There's certainly a sweet spot between the high end prices and the China town price, as there is with everything. The former means you're probably being exploited, the latter means someone else is being exploited and you're getting a piece of shit as a present for your low price.

Quote: Originally posted by Crucible

@peach I don't know that I'll ever get my plate up to 1000C, but I designed it so that I would have the option. I'm a big fan of having few pieces of equipment that have a wide range of capabilities. Sorry if that confused you.

I did do a fair amount of complaining. But, now I know a bit more about what you're trying to do, I'd say it's a fair attempt. Don't think I'm getting all nice on ya, that's never going to reach 1000C.

Not because you've built it poorly, but creating a large surface area exposed to the atmosphere and electrically heated to 1kC isn't so easy.

At first, I thought you were up to something pyrotechnical or highly explosive with it.

Quote: Originally posted by Crucible

My main concern with explosions was indeed volatile solvents (ether, petrol distillates, etc). And my concern with that is joints leaking. Having some fumes leak, ignite, and flash back shattering all my precious glassware, spread flaming solvent everywhere, and shower me with broken glass would really ruin my day. I always try to prepare for the worst case scenario, so that's why I went for explosion proof. Isn't that what it's for?

If yar be usin' jointed glassware, grease it and add keck clips. I know you're trying to cover all the possibilities with the plate, but it might be causing you to overlook the more simple.

I've distilled lots of flammable solvents and things like nitromethane, and I smoke; sometimes stood next to it. It'll take maybe two or three minutes to grease and clip the joints and there's no way enough vapour will escape from those to ignite. I could hold a lighter against the joints and not see anything.

If you're worried about flashpoints and autoignition, reduced pressure distillation will lower the BP, so it'll be away from those temperatures. Flooding it with N2 first will mean there's essentially no way the fuel can ignite, because the amount of oxygen present will be so far below the ignition requirements and the volume will be saturated with fuel (which also won't burn in high mixes). If you reduce the pressure enough, you won't even need to synthetically heat the flask, it'll boil as it absorbs heat from the air around it.

It's a good idea to be safe, but those joints and exhausts are by far and way the first port of call.

Saying all this BUY A BIG FIRE EXTINGUISHER AND HAVE A HOSE HANDY

I've had to extinguish someone who set themselves and a plastic tank alight with petrol. The screaming people make is instantly recognisable from hollywood or anything else, it is animalistic and communicates pure fear. And every fraction can make the damage exponentially, irreversibly worse.

He was using the petrol to light a BBQ, and all I had to hand was a small kitchen powder extinguisher. It ran out in seconds, and I had to fill buckets of water to put the tank and garden out. When people are on fire, the most disheartening thing you'll feel is the extinguisher running out.

The tanks are designed not to flashback, it's not like in the films when they explode.

And, luckily, he'd only set his feet on fire. He had some blisters and the marks stayed for a few months, but they got better.

I laugh remembering him running around with his feet on fire, but I still graphically remember the noise, the fear and adrenaline. It would be absolutely horrible to have that happen and for it to be worse, with no extinguisher.

He set himself on fire after managing to ignite the fuel tank's open lid. Thinking it'd explode, he'd kicked it to try and put it out. Where upon the petrol sloshed out all over the floor and his feet.

A prime example of panic making things worse.

I would also draw a conclusion from this that you are most likely to set yourself a light when you're handling the flammable solvents at the start or emptying the still, rather than a freak accident during distillation.

Let it cool right back to room temperature. Before opening the joints.

If the glass ignites, the flame will burn outside the glass where the oxygen is, not inside. The biggest risk is that the glass shatters and empties the fuel onto the floor in a gigantic puddle. To this end, put the glass in a big aluminium pan that'll catch it if it tries to escape. This will be very easy to control, low risk and extinguish by comparison. You can by them cheaply from commercial food preparation warehouses, they sell them for making big trays of dinner for people at school and in work.

A big tank of liquid CO2 costs next to nothing. They are wired to plastic hoses that burst when a fire breaks out in commercial, high risk areas.