How to make dosing pump system more accurate

Thread Starter

james211

Joined May 29, 2012
267
So I have an inexpensive dosing pump system I purchased, its an inexpensive chinese product. The software is OK, and the pumps were mediocre at best. I had four welco pumps here at home which are very well made japanese peristaltic pumps so I thought I'd swap those into this unit. The pumps run great but even with the calibration in the software, they aren't as accurate as I'd like. So my question is, is it possible to put a very fine tuning potentiometer between the board connection and the motor so I can fine tune the speed of the motor for more accurate dosing?

Here is the doser: I Paid $60
http://www.fish-street.com/jebao_auto_dosing_pump_dp-4

Here are the pumps I put in place of the stock pumps: I have the WPM1-P2BA-WP
www.welco.net/dcms_media/other/WPM_selectguid.pdf

Any assistance would be awesome...thanks again.
 

AnalogKid

Joined Aug 1, 2013
8,934
A lot depends on the voltage across and current through the motor. Can you measure either one of these? Overall, you probably will not be able to find a multi-turn or other high precision pot that will handle the motor current. Another way to go is a low dropout voltage regulator, which can be very precise if needed.

ak
 

MaxHeadRoom

Joined Jul 18, 2013
22,341
Generally peristaltic pumps are fairly accurate/rev, and very low rpm, how are the motors/dosing controlled now?
If it is a question of precise rev count, it may be possible to use a small prox detector.
Max.
 

crutschow

Joined Mar 14, 2008
26,361
I would think that counting the revolutions as Max suggested, would give the most accurate control of the dose.
That makes the accuracy independent of the motor speed.
 

KMoffett

Joined Dec 19, 2007
2,918
If you were going with an all new controller I would suggest the pump head with the stepper motor. That would give you the most accurate rates control. Also a controller with a standard brushed motor/PWM, and a rotation counter would work.

Ken
 

MaxHeadRoom

Joined Jul 18, 2013
22,341
I don't think you would even need a PWM control with a DC brushed as these dosing pumps often run around 3rpm and stop fairly quickly due to the gearing, which is often enough, use a rev/hall effect counter if needed.
Max.
 

Thread Starter

james211

Joined May 29, 2012
267
Whoa...I didn't receive notification of any of these messages, so here a I am. For starters, this unit has a complete control system built into it, so my guess is either I build a new system from the gound up, or I figure out a way to reduce the RPM's of these pumps. To answer a a few questions, as you guys clearly have more knowledge than I do:

1 - These pumps definitely run at a higher RPM rate than 3rpm's. Tomorrow I can give you a more accurate dosing rate, but they are definitely faster than that.

2- These pump are brush motors

3- Not sure if the system runs on PWM, seems they run at full voltage ( I can check this out later as well).

4- Ideally I'd use a stepper motor, but I have these Welco pumps, which like I said, are pretty much top of the line. you cannot just order these brand new. I had to do a lot of talking to get them into my hands.

5- Keep in mind, I have very limited programming experience, hence me asking about a potentiometer. I'm sure you guys dislike idiots on this forum, but thats why I'm here.

Questions on my end....is there any simple way to check if its using PWM? Again, I'm guessing it doesn't based on the RPM rate. To me it seems more like an "on-off" kind of situation.

Once again, thank you for your help.
 
Last edited:

Thread Starter

james211

Joined May 29, 2012
267
The calibration is run via a program function that counts, in seconds, how long it takes to pump 100mL of fluid. The software then converts that information into how many mL/sec. What I don't like is that you cannot manually tweak that number.

Just out of curiosity, what would happen if one was to put a potentiometer between the board and the motor?
 

ronv

Joined Nov 12, 2008
3,770
The calibration is run via a program function that counts, in seconds, how long it takes to pump 100mL of fluid. The software then converts that information into how many mL/sec. What I don't like is that you cannot manually tweak that number.

Just out of curiosity, what would happen if one was to put a potentiometer between the board and the motor?
Seems like it should work pretty good. I think I would have counted revolutions, but hey. :rolleyes:
Depending on how much current the motors take the potentiometer would probably smoke. And motors sometimes have a hard time starting with resistance in series with them as the start up current is higher than the run current. What you might try if you want to slow it down is a diode in series with the motor.
 

Thread Starter

james211

Joined May 29, 2012
267
You know what, I spaced, when I was referring to a potentiometer, I was referring to these mini PWM controllers with a pot on it. My apologies, I dabble in this stuff.

So given that the computer turns on each motor port and basically just supplies power, I could in theory use that as the power supply to the PWM controller, and then plug the motor into the controller and slow it down that way?
 

MaxHeadRoom

Joined Jul 18, 2013
22,341
As per post #4 #7 #12.
They work pretty slick, I picked one up for evaluation.
They are a simple circuit based around the 555, but for $5 it is hardly worth the effort of building?
Max.
 

mcgyvr

Joined Oct 15, 2009
5,394
Don't forget it might not be an electrical issue at all.
I wouldn't be surprised if the inaccuracy came from more than just motor speed/control timing..

It could be from mechanical aspects too.. The mechanical linkage from motor to shaft is hardly reliable and relies on friction only.
At the speeds that motor turns it could "easily" be slipping sometimes..
Sealing of the silicone tube could be an issue thus loosing its "vacuum"
and many more reasons..

I personally have one on my reef tank and have NO problems with the accuracy... Don't know how accurate/repeatable it really is but it controls my alkalinity/calcium/carbon dosing (vinegar) just fine without me having to mess/adjust it
 

Thread Starter

james211

Joined May 29, 2012
267
are you talking about the shaft into the rollers? The pump I have actually has a geared shaft so it makes a direct connection with the rollers. That's part of the reason I don't like the inexpensive pumps. It seems weird to me that anyone would rely on a pump that has no geared connection between the rollers and the shaft.
 
Top