Sciencemadness Discussion Board

Building a peristaltic pump for small volumes/rates.

Fulmen - 4-1-2015 at 09:41

Figured this could require a separate thread, it's for dosing HCl in a chlorate cell (main thread here: http://www.sciencemadness.org/talk/viewthread.php?tid=53350)

The cell will draw up to 8A, and the most quoted consumption rates is appr 0,15ml 12% HCl per A per hour giving a ballpark rate of 1ml/hour. The actual rate and HCl-concentration will have to be determined experimentally, so I will need a fairly broad working range.

This is a rough sketch of the design I'm considering:
Pumpe.jpg - 18kB
A 3mm ID PVC-tube and a 50mm bore happens to equate to roughly 1ml per revolution, seems like a good choice to me. As for motor and driver, I'm not sure which route to go...
I can get geared DC-motors down to single digit RPM speeds, these can be controlled by a combination of speed control and timer, or perhaps with a monostable vibrator and a self-parking design (think windshield wiper) and a timer.
The other idea is to use a geared down stepper and a pulse generator. I've found 2000-6000step motors delivering a few kg-cm of torque that seems interesting, but I'm uncertain about the required torque.

I will of course have to build a prototype somewhere down the line, but the more I can figure out before I start machining the better.

And yes, I know I can buy these things on ebay, but complete units are too expensive for me right now.

YourUncleJoe - 4-1-2015 at 10:50

While I can't speak to your torque requirements or the mechanical design, I can help with the electronics. Stepper motors are fairly straightforward to use. Pololu electronics has a nice selection of motors of varying sizes (NEMA 13-17, if I recall correctly). Here's a nice introductory lecture on stepper motors, if you're curious. Driver circuits for them can be purchased fairly cheaply, and they're easy enough to interface with something like an arduino. How much electronics and/or programming experience do you have?

Fulmen - 4-1-2015 at 11:07

Programming: Not much, nor do I have any interest in it. So Arduino-style controllers are out. I have built some circuits using simple ICs and discrete components, don't expect much problems whipping up something like a monostable multivibrator or a pulse generator.
I do like steppers, and since you can buy ready-made drivers for scraps from DX or Ebay I am seriously considering this option. This would also provide continuous delivery, and while it isn't necessary for this project it could come in handy for the future.

macckone - 4-1-2015 at 19:23

Sometimes it is better not to reinvent the wheel.

http://www.adafruit.com/products/1150

j_sum1 - 4-1-2015 at 22:21

Quote: Originally posted by macckone  
Sometimes it is better not to reinvent the wheel.

http://www.adafruit.com/products/1150

I was going to say the same thing.
http://www.ebay.com.au/itm/DC-12V-Dosing-pump-Peristaltic-do...

Aquarium suppliers have products that do exactly what you want. I bought a model similar or identical to this one a couple of months ago but haven't used it yet.

Fulmen - 5-1-2015 at 03:52

Fair point, but I've already made a decision. Besides I can build the pump in a fraction of the shipping time, in fact the housing is already roughed out.

Here's the design so far:
Pumpe2.jpg - 32kB

Question: are there any real benefits to more rollers? Less pulsing perhaps (not an issue for me), but other than that?

Fulmen - 9-1-2015 at 08:59

The pump body about 2/3 done:
Proto1.jpg - 451kB

Still need to mill ports for the tube and make a lid, and of course make a decision on the motor. I've pretty much narrowed down the choice to these two:
http://www.dx.com/p/15mm-2-phase-4-line-micro-15by-dc-5v-ste...
http://www.dx.com/p/zndiy-bry-12v-dc-3rpm-torque-gear-motor-...

The 5mm PVC-tube I've hoped to use is quite stiff, I'll try to jerry-rig a way to measure the required torque. But in worst case I'll just get some silicone tubing.

As for the motor, can anyone come up with a good reason for NOT choosing a stepper?

smaerd - 9-1-2015 at 11:16

Looks great. How did you machine the pump head? I have plans for building a peristaltic that I threw to the wayside due to lack of available parts. After doing a ton of research I settled on tygoprene tubing I think it was for the pump. Tubing is important, which I'm sure you (op) know but for a passer-by, hose can only be pinched&rolled so many times before a failure especially in the presence of solvents/chemicals.Not sure what PVC's resistance to 12% HCl is, but I'll assume you did your research.

Stepper would be the way to go imo. That way your peristaltic pump isn't a one trick pony(can always adjust the speed 'reliably'). NEMA 17's or whatever you want. The ones you have linked look fine just keep in mind the load and the RPMs/flow you want. I recently ran into an issue with using a L293D driver IC on a 1A stepper. The IC get's scorching hot even with a heat sink. Misleading spec sheets!

Magpie - 9-1-2015 at 11:29

This is a nice project Fulmen. For one experiment I needed a small peristaltic pump for a synthesis but nothing was available except expensive medical or industrial pumps like a Masterflex.

I have some industrial experience with a peristaltic pump. It was used to pump a hot, viscous, radioactive, salt brine continuously for days on end. We found that the originally specified hose (EPDM, IIRC) would fail after a few days. We replaced this with a hose made of polyurethane. This would last for several weeks before failure. In fact we just replaced it on a schedule and never did again have a failure.

How will you control the pump speed? Will you use a PWM electronic driver like I have done with my stepper motor overhead stirrer?

Fulmen - 9-1-2015 at 11:30

Quote: Originally posted by smaerd  
but I'll assume you did your research.


Looking behind me while whistling *_*
Nah, just kidding, PVC should be fine for HCl. I'm more worried about the mechanical wear, although this pump won't turn more than a few dozen times a day. So even a less-than-ideal tubing should last a good while I hope. Not that I'm married to the PCV, it's just the simplest to get hold of.

As for the body it's cut on a lathe from 6063T6, if it works properly I'll anodize it for looks and durability.

macckone - 9-1-2015 at 15:32

PVC is good for most inorganic acids.

Fulmen - 10-1-2015 at 04:07

Huh. I thought I knew what I needed to control a stepper, but looking through the circuits at DX.com (my favored shop for such things) I'm getting quite confused. Only worked with a CNC with higher-end controllers in the past, these were integrated drivers/controllers that had step/dir inputs. A lot of the circuits has 4-pin inputs, I'm assuming these are simple H-bridge drivers that require a separate controller? Hard to tell what is what from the descriptions, so I'm left to guess from the pics.

This seems to be a complete driver/controller: http://www.dx.com/p/tb6560-3a-single-axis-controller-stepper...
3A might be overkill, but it's a very versatile circuit that could be used for larger steppers as well.

As for the motor (http://www.dx.com/p/15mm-2-phase-4-line-micro-15by-dc-5v-ste...) I'm assuming it's rated for 400mA (calculated from 12V/30ohm) even though the title says 5V.

j_sum1 - 10-1-2015 at 04:17

diddi and magpie recently had a detailed technical discussion on using a stepper for a stirrer. You might want to look up that thread and get some details.

Edit
Here is the thread.
https://www.sciencemadness.org/whisper/viewthread.php?tid=41...

[Edited on 10-1-2015 by j_sum1]

Fulmen - 10-1-2015 at 06:21

Thanks. I think I found what I was looking for by searching for the chip: http://www.dx.com/p/hf-a4988-stepper-motor-driver-module-for...

This should be more than powerful enough for my use and cheap enough to buy a handful. Always sensible to have a few spares for experimentation.

Magpie - 21-1-2015 at 12:49

Fulmen I hope your pump head construction is proceeding well. I have some questions pertaining to such a small peristaltic pump. Although I will not be building my own pump head, I thought this thread would be a good place to ask.

I would like to have such a pump for various experiments. I specifically need one when making CCl4 from CHCl3. In this case I need to dose TCCA with dilute HCl at a steady 0.5 drop/s. This low flow is very hard to produce with a p-e funnel. This requires close monitoring for up to 10 hours. With a reliable dosing pump it would take much less supervision.

My plan is to buy either the small dosing pump offered by adafruit, as linked above by macckone, http://www.adafruit.com/products/1150, or what appears to be an equivalent (but cheaper) pump from China as linked by j_sum1, http://www.ebay.com.au/itm/DC-12V-Dosing-pump-Peristaltic-do...

Adafruit says that their dosing pump can be controlled by PWM using motor driver L293D.

I have two questions concerning this approach:

1. Is the cheaper Chinese pump equivalent to that shown by adafruit?

2. What would be an appropriate electronic circuit for controlling such a pump by PWM?

edit: the second link above seems to be defunct. Here's an equivalent: http://www.ebay.com/itm/12V-DC-DIY-Dosing-pump-Peristaltic-d...


[Edited on 21-1-2015 by Magpie]

[Edited on 21-1-2015 by Magpie]

[Edited on 21-1-2015 by Magpie]

Magpie - 31-1-2015 at 16:29

A 12vdc peristaltic pump I ordered from China arrived today. I connected it to a motor control circuit to see how it would perform. The YouTube link below shows the pump in operation.

The control circuit utilizes a 555 timer and a transistor to provide pulse width modulation (PWM) of 12vdc. Parts for this came to about $12. The pump itself was $11 from China. 25' of tubing cost $5 (McMaster-Carr).

http://youtu.be/jZZXV3DuTc4

The control circuit is the one shown at this website:
http://www.electronics-tutorials.ws/blog/pulse-width-modulat...

If you have any questions please ask.

Zombie - 31-1-2015 at 19:28

That's a cool set up.

There is a whole thread here on stepper motors / timing controllers that has all the info needed for one of these builds.

http://www.sciencemadness.org/talk/viewthread.php?tid=41539

It may help with the low rpm delivery issue.

smaerd - 31-1-2015 at 19:35

Nice video magpie! Happy to see that you got it working.

Magpie - 31-1-2015 at 21:35

Quote: Originally posted by Zombie  

It may help with the low rpm delivery issue.


Thank you.

The low rpm issue is a characteristic of the motor/pump head, not the controller, I believe. I suspected I might have this problem but thought I'll never know without trying it. Using the adjustable clamp on the outlet is an entirely satisfactory solution for this problem. My target flowrate was 30d/min. This was just a little too low for this pump without the outlet clamp. For the higher flow rates it works just fine using only the variable resistor (potentiometer) for flowrate control.

The pump motor itself is running at 5000 rpm. So I believe that there's already considerable speed reduction using gears.

The pump comes with a short (4") section of silicone tubing. I removed this and installed PVC tubing having a slightly smaller OD but a slightly larger ID. I needed something that would be compatible with 20% HCl. Luckily PVC is much cheaper than silicone.

Thanks smaerd. ;)

jock88 - 1-2-2015 at 12:21


I have seen very small peristaltic in some ink jet printers for sucking ink out of the print head (cleaning routine). They may be available from dumsters etc

Magpie - 11-2-2015 at 11:54

Here's a picture of my completed variable speed peristaltic pump. The project box is a duplex outlet box with cover ($2) put to a higher purpose.

Only the PVC tubing will be in my hood while the pump is in use. The tubing just happens to conveniently fit through my existing wall penetrations.



peristaltic pump station - side view.jpg - 93kB

Fulmen - 11-2-2015 at 12:07

Nice work. Sadly this project has been sidetracked by other more pressing work, and I fear it has lost it's momentum completely. It didn't help that the stepper I bought was a bit of a miss.

Not sure where to go from here, since I have to order new parts anyway I might end up buying a cheap unit and jerry-rig a timer for intermittent dosing. That way I can get on with the main goal and finish the homemade pump at a later time. I still want to complete it and make an accurate, all-purpose pump with a wide working range, but I can live with a simpler solution for now...

smaerd - 11-2-2015 at 14:20

What was wrong with the stepper?

It's kinda sad to see it go because you have such a nice pump head already. I definitely understand when projects get cancelled though.

Zombie - 11-2-2015 at 16:05

Viery nice, and compact.
You could mount the entire assembly inside another 8 dollar "hobby box", and have a real "pro" looking rig.
Just 2 cents...

In fact I'd recommend a metal box, and earth grounding the whole deal thru a GFIC outlet.
2 more cents...

Fulmen - 11-2-2015 at 16:44

It was just too small, even if it was strong enough it would be hard to attach it to the pump properly. And I'm not throwing it out, just putting it on hold for a while.

Magpie - 15-9-2018 at 17:34

A week ago I had three peristaltic pumps that didn't work. Two of them were brand new. Today all 3 work. For the two new ones the silicone tubing had taken a set and was closed off. For the old one the PVC tubing was too small.



3 peristaltic pumps.JPG - 2.3MB

DrScrabs - 17-9-2018 at 02:13

I did not read all above so sry if my post is redundant
DIY peristatic pump with stepper motor + code