Sciencemadness Discussion Board

Voltage varying centrifuge problem

resveratrol - 21-9-2011 at 18:09

My centrifuge is 3365 RPM, the radius of the rotor is 12.5cm (factory default) operates at 115V....Need to figure out what I have to drop the voltage to for the centrifuge to give me 72 g's

not_important - 21-9-2011 at 20:57

Well, what rotational speed do you need for that g force? Wiki has the formula you need, I'd bet.

note that the type of motor used is quite important as several common types are slaved to the excitation voltage frequency (read: power line frequency, 60 Hz by your 115V comment). With those types you aren't going to have much effect on the rotational rate by messing with the voltage, until you reach a point that the motor is malfunctioning.

peach - 22-9-2011 at 02:58

For 72g's at a 125mm radius, you need a rotor speed of 717 rpm.

The default RPM is 3365, so 3365 / 717 = 4.693165969

The default voltage is 110, so 110 / 4.693165969 = 23.44V

To adjust the AC mains voltage, you will want a variac;

You can get a more accurate reading of the voltage, for repeatable results, with a cheap multimeter;

To check the rotor RPM, you will want an adjustable strobe;

You should not try fixing any form of tag to the rotor (such as a magnet to measure the RPM's with a hall sensor) as you will unbalance the mass, which can have catastrophic effects when the rotor spins up and then breaks free. If the rotor gets you in the face, there is a good chance it will kill you. Modern centrifuges can store as much energy in the rotor as there is a stick of dynamite;

I am assuming you are doing this on something of a budget, as a university, school or health body would just buy an adjustable centrifuge. With that in mind;

- Determine roughly what the power rating of the motor is, so you can buy a suitably sized variac. The big ones can cost a lot of money, but the motor may only be a few hundred watts if it's a small centrifuge. Variacs without lots of switches and meters built into them are cheaper.

- Get a cheap multimeter and use it to set the voltage.

- Don't get the strobe. Unless you're good with electronics and feel the urge to build your own (an LED being turned on and off by a pulse generator will work, provided you can measure the pulse frequency <--- oH-silly-scope), buying them new can be a bit expensive. Instead, set the voltage, let it spin and then adjust it based on the results you get out. In terms of biology, the variation in the samples going in will mean even if you set it at precisely 72g's and run it for precisely X number of seconds, you may not get precisely the same results out as you're being quoted anyway, and will have to let it go a bit longer or for less time. So being super price with the RPMs and duration may not be worth the cost of the strobe.

[Edited on 22-9-2011 by peach]

resveratrol - 22-9-2011 at 05:50

This is my centrifuge

@not_important you might be right, and I was worried that I might be limited by something like that

@peach wow, thanks. a few things though:

"3365 RPM / 717 RPM to yield a factor of 4.69" this is exactly what I was thinking, but I was unsure as to whether or not you could simply divide the voltage by 4.69 and get that result. It sounds good on paper, but I just wanted to make sure I didn't have to account for gravity somehow.

and I don't think the strobe will be necessary. The 3365 RPM is not factory default, that's what the factory calibration sticker says on the back of the unit (3,380 is default) - not to mention it has very minimal use and is practically brand new. So ideally it should work just as well as my math does lol.

And do you think something like this would be reliable?


not_important - 22-9-2011 at 07:06

Spec sheet sez "brushless AC motor", rather good chance of it being locked to and designed for the power line frequence; meaning if so you're SOL.

That Ebay item specifies AC/DC motors with brushes, which can have their speed controlled by the voltage applied.

Why would you want a lower G, generally that's counterproductive, As for gravity, that's only a bit over 1% of your target Gs, almost in the noise and likely too small to matter or be controlled for - someone in the building turning on their blowdryer would wiggle the line voltage that much. Besides, it's a horizontal unit so the Earth's G field is at right angles to the `fuge's force.

resveratrol - 22-9-2011 at 07:32



and you're right about gravity heh, not sure what I was thinking. The lower G is because its for biological separations (i.e. PRP preparations). A higher G would shove the platelet-rich region (which is what I'm after) right down with the red blood cells.

Hmmm...what would you do then?

The WiZard is In - 22-9-2011 at 11:01

Quote: Originally posted by not_important  
Spec sheet sez "brushless AC motor", rather good chance of it being locked to and designed for the power line frequence; meaning if so you're SOL.

That Ebay item specifies AC/DC motors with brushes, which can have their speed controlled by the voltage applied.

Yup. However, if you are willing to spend the money...
you can buy an AC motor speed controler. It may be
less expensive to install a DC then you can vary its speed
by changing the voltage. This is assuming that your
motor will work with a phase motor speed control!

Google it.

peach - 22-9-2011 at 14:55

I was worried that'd be the case, as not_important has suggested.

You may need a variable frequency drive to control the RPMs. They're available on eGimp starting at about £80.

You'd need to be careful swapping the motor to ensure all the load carrying surfaces, fixings and balancing remained suitable. It is a possibility, but it'd be safer and cost about the same to get a VFD.

A higher G would shove the platelet-rich region (which is what I'm after) right down with the red blood cells.

Hmmm...what would you do then?

Run the cycle for less time? :D

[Edited on 22-9-2011 by peach]

resveratrol - 22-9-2011 at 15:23

Yeah I was looking into those variable frequency drives, but shit....for that price I could buy another centrifuge

And running the cycle for less time was the first thing I thought of, but that would give me worse resolution dont you think?

peach - 22-9-2011 at 16:19

It may slightly, as the particles will have less time to settle in an orderly manner. But it's a question of convenience over perfection. You could leave the sample in microgravity at absolute zero and it'd give a nice sediment, but take ages. See if you can get workable results from the centrifuge without messing around with the RPMs first.

The best plan would be to give a test spin.

A butcher can provide you with a cup of sample juice.

Alternatively, sell the bastard and buy a different one.


Nah, I'm only kidding, I'd never suggest you buy a Dyson. You need a Bosch. :P

{edit}Imagine how many vials you could get in all those holes!

[Edited on 23-9-2011 by peach]

resveratrol - 22-9-2011 at 16:36

Lol, you are silly!

if only there were a simple solution around this problem. maybe if I turn my centrifuge off and on really really fast that will give me low enough G's.

in all seriousness though...I don't think it's worth trying to reduce the spin you really think I could get workable results?

magnus454 - 22-9-2011 at 22:05

Whats the web address for eGimp? searching just takes me to sites involving the Linux image editor gimp. Shorter time on high speed will give you lower resolution, long on a lower speed will give better resolution.

Twospoons - 22-9-2011 at 22:25

Its only a low power motor, so a cycle skipping controller might do it. Don't know where you'd get one though.

resveratrol - 23-9-2011 at 13:04

So if 15mins @ 717 RPM is ideal....that's:

(717rpm) x (15min) = 10,755 rotations

(3365rpm) X = 10,755. X =3.196

So that's going from a spin cycle of 72 Gs for 15mins to like 1581 Gs for 3min 11sec

Do you think my resolution will suffer? Provided that is even the correct math...

resveratrol - 4-10-2011 at 16:23

So...I emailed the manufacturer. Turns out I can trade my unit in for a refurbished variable speed model (which retails normally for $970). It will cost me $200 for this service, but I think it's totally worth it. =)