The rebirth of the peristalic pump?

I was reading the archive on the peristalic pumps and saw that the major issue was the pulseing flow of the pumps. Might I suggest linear peristaltic pumps, since the shoes are identical (thus easy to reprap and replace) and it doesnt pulse like a circular peristalic. The only difficult part to make would be the S-shaped axis in the center. The use of a spring loaded pressure plate on the bottom would also make looser tolerances more feasible.

[chlauchpumpe-lineare-Bauart-transparant-indizert-bewegt.gif" target="_blank" rel="nofollow">en.wikipedia.org]

Mike

The thoughts and ideas expressed in this post do not reflect those of my employer and are intended only as communications between individuals. Any attempts at implement are at your own risk

Edited 1 time(s). Last edit at 09/12/2007 09:44PM by ohiomike.

The link was messed up but I found it. [upload.wikimedia.org] Thats pretty cool. Must be lots of wear and tear on the tube.

There's only up and down pressure on the tube, from the row of rectangular plates. (The rectangular plates only oscillate in the z axis.)

Nice.. sure...

But i guess pretty complicated and expensive to build one of these..
You need a crank shaft, some .. how do you say Kipphebel? rocking lever??
and the pushers.. and of course a bed for the tubing...

that's why I guess it'll be easier to build one of those extruders
[reprap.org]

or a pump like that one I do not fnd anymore
(I'm sorry, but a search for the wiki could be quit useful )
basically one turning rocking lever(I'm pretty sure that this IS the WRONG term)
only

'sid

i think something like this would be far easier to build:

[www.eccentricpumps.com]

less pieces involved. but i'm just not quite sure any of these persitaltic pumps can deliver a steady enough flow

Thats a sweet design. Simpler and more efficient

Wow! That's a BEAUTIFUL little design. I bet I could power one of those with a GM2.

Now that I have Tommelise printing I think I'm going to try to build one of those for extruding spackling compound for support material for Tommelise!

Thanks for the link, Fernando! I think you may have just saved me a bunch of design time.

Glad I could be of assistance, Forrest!
It seems that i have a thing for finding documentation

here, this actually answers my doubts about this design being able to deliver non-pulsing material flows:

[www.eccentricpumps.com]

As it seems the pump-design is pretty steady per se and if you add an accumulating vessel behind the pump, the flow steadies even more. To achieve this, they have a picture in the document above. It involves having a stand pipe at the exit of the pump. I imagine that's just a piece of pipe open (?) at the upper end, which dampens any surges in the flow.

I thought the pump would be integral to the delivery system. How does the accumulating vessel figure into that?

Well in the link i gave above you get a pdf that has a picture of the pump device they built. it was made to test the flow of an adhesive, so it really relates to what we are trying to do.

I have attached a picture of what i think they have done in the experiment rig to get constant flow. The question that i don't have so clear is whether the standing pipe is open or closed at the top and how long the pipe has to be, but that's easy to test and find out.

i agree... very cool design. it would be pretty simple to reprap too. just a housing body w/ grooves for the tubing, mount for the motor, and the motor arm piece. it looks like there is a wheel on the end of the motor arm... if you need a bearing, a $1.00 skate bearing would probably work great! you could also slap a reprapped 'wheel' on it to get a larger/wider wheel.

neato!

I think that the accumulating vessel was only pertinent to what they were doing in recirculation. In our case, we would have the pump directly providing pressure to the extrusion orifice. It would be a matter of matching the speed of the pump motor to the extrusion rate needed, much like our mk2 extruder does with the filament melted and squirted out the end.

I'm sorry..

I was sure one of those eccentric pumps was already reprapped !?
I can't find that anymore so maybe I just grow nuts

'sid

bartlee45: dunno, you may be right, i'm not a pump expert
What you say about the flow rate is true. we just need this pump to work at pretty slow pump rates. Is 1ml per second an accurate estimate or is it below? It would be easy to make a small calculation and see what sizes and rotation speeds we need.

Zach: As pump designs go, this must be the easiest and cheapest to reprap. Good idea about the skateboard tyre!! Does anybody have an idea what type of tubing these pumps work with?

hmmm perfect pump:
[www.mahr.com]

but expensive

'sid

sorry for advertising expensive resinpumps...

yes, adrian did some research on peristalic pumps within the last year. if you go onto the blog and search, i'm sure you'll find some info on it. i believe he made the pump parts via RP.

The problem with the regular peristalic was that the high viscosity of the material meant that bubbles in the line were a problem. Filled polymer systems such as the polyfill (and my experiments with fillers) are almost always dialent suspensions and thus difficult to pump. My hope was that the large surface area of the linear peristalic would increase the vaccum enough to prevent so many bubbles from forming.

I am currently working on a two part polyceramic composite that should be pumpable with the orginal pump design, assuming that the supply tanks are kept aggitated.

Mike

The thoughts and ideas expressed in this post do not reflect those of my employer and are intended only as communications between individuals. Any attempts at implement are at your own risk

Edited 1 time(s). Last edit at 09/12/2007 09:44PM by ohiomike.

Aha, i see the problem...
We have to solve it somehow, pronto.
We MUST have a method to pump high viscosity polymers with fillers at a constant flow-rate!

I'll be giving it some thought.
Right now, the only simple solution i see is to use piston pumps, like my cartridge idea. If we can have a refill cycle (just like a gas engine refilling with fresh gasoline-air mix) refilling the cartridge with polymer blend from a big, stirred polymer tank, we could print large quantities with regular flow, no bubbles.

Question is, can piston pumps deal with the pressures involved with high viscosity polymer blends? Also, to keep the system flowing, the curing of the polymer can only start after extrusion from the nozzle, or else the whole rig will become clogged at some time.

Piston pumps are actually very popular for high viscosity pumping, pound for pound they put out more PSI than any other small pump design. I used to use one to pump a paint concentrate (that could best be described as sludge) up two stories and across an unheated warehouse, in Ohio, in January. They were metering pumps manufactured by LMI.

But that would mean we need to be able to build checkvalves that can take the pressure, or we need to manually switch the valves from extrude to refill. The advantage of course would be that with checkvalves the stroke of your piston no longer determines the volume of your system but only the maximum possible pressure.

Fisher Scientific has checkvalves available, $180 for a box of 50, I am looking to see if I can find a supplier that has them as single units.

[new.fishersci.com]

As far as blending the material, thats less a problem then it might seem, a lot of paint systems inject the catayst just before the material is applied so while it would require the use of two pistons or two pressure pots (or maybe one of each since cataysts are usally very small additions compared to the resin) it allows you to have the materials activate within seconds of being applied, which allows you to pick cataysts that harden quickly and get very high productivity.

Mike

The thoughts and ideas expressed in this post do not reflect those of my employer and are intended only as communications between individuals. Any attempts at implement are at your own risk

Edited 1 time(s). Last edit at 09/12/2007 09:46PM by ohiomike.

Yeah, i think the piston pump is a very promising solution.
It allows to be scaled to suit everybody's needs: from the syringe through the cartridge, rechargeable cartridge, and continuous pump for very big printout quantities.

I will probably be using cartridges up to 600 grams (from here, they have the cylinders, plungers, mixers, everything. and pretty cheap too. [www.cammda.com]). Those should suit my needs. But the check valves that you showed here are the solution to make it refillable. I would say that stopping the print process every time you need to refill the cartridge is probably the easiest and best solution. I guess I could find a check valve in my local plumbing shop for little money.

The issue about mixing in a catalyzer at the level of the nozzle would be a great advancement. In the link i give above they have helico

I'd stick with the peristalic pump, myself. If it has a pulsing flow couldn't you control the motor speed in real time to smooth that out? We've got a lot of experience with gearmotor control via microcontrollers. It shouldn't be that bit a deal or that expensive it it can be done at all.

also, hopefully you would be able to print that nozzle that needs replaced. then you just print a batch of 40 of them, or you could just have it print a new nozzle as part of each print run.

That's what I plan to try, anyway.

Forrest: The peristaltic pump is a fantastic solution for single component, low viscosity polymer. Replacing the tube when necessary is pretty easy and cheap. But you won't be able to print high viscosity fluids, which means no polymer blends with heavy filler loads. That's to bad because it's a fantastic material to print big, chunky durable and solid parts, that can be worked with after setting.
I think we have to have a palette of solutions that would allow us to print out different types of parts.

Zack: Indeed, I have an idea about mixer nozzles that could be printed. It's just a matter of having a long slender tube with obstacles along its way, so that the turbulences created in the polymer blend while traveling through it would mix it. No need of it being helico

The Mk 2 and its descendants do a fair job of printing high viscosity polymers already. I don't need another way to do that just now. Printing support material like Pollyfilla (spackling compound) is what I have in mind.

i have found the documentation of this dispenser: [www.reprap.org]
Is that the mk2 one you are talking about? i thought it was aimed at extruding thermoplasts only. should i understand that you tried with high viscosity non-thermoplasts and got good results?

negative. that is a thermoplast (or anything that goes from solid -> viscous < 200deg C) extruder.

what forrest was referencing is the syringe extruder design. i dont know whats up with its documentation, but there should be something around on the website about it. that is the one that is for pumping non-thermoplast (such as support filler).

***what forrest was referencing is the syringe extruder design. ***

LOL! No Forrest wasn't. He just likes filament thermoplastics rather than thermosets or epoxy-like plastics.

I was thinking that if you want to print some heavy duty stuff, just reprap a thermoplastic mold and then fill it with whatever you want. You might be able to reuse it. Then there is no fancy mixing, just a stirring stick and some disposable containers to mix in. I'm sure that many materials will fall into the hard to dispense but great to be made out of category.

bartlee45: point taken! I prefer making molds too.
Some people may want to have heavy duty printers (i can think of people wanting to print furniture or such), but that application may be further ahead in time. for now the mold-printing solution is simpler.

Forrest: i think that having some sort of support material dispenser is very important too, i wasn't saying we shouldn't do that. actually for printing molds you kind of have to have something of that sort to make all the complicated volumes that you'd need. But I haven't yet figured what whay to go: powder? high density liquid? waxes? water soluble waxes or gelatine? I think waxes are the easiest, as you may use your thermoplast extruders to do the task: just find the good temperature to melt stearate paraffin (very cheap and melts at 68