I need to control a stepper motor which is driving a high pressure pump in such a way so as to operate it in two modes: constant pressure or constant flow rates. The stepper motor in question is a four phase using two phase exciting system, with max motor current around 5 A.
I need ideas about how to go about accomplishing this and whether an arduino uno can be used along with some off the shelf board to control this system.
I am a biochemist so I not very proficient with nitty gritties of electronics, and hence I am trying to look for simple elegant solution to power this stepper motor which can power my pump for the HPLC experiments.
First you need to get a suitable stepper driver for your motor - one that can comfortably supply 5 amps.
Then you need a suitable power supply.
It would be useful if you provide a link to the datasheet for your motor.
application...application...application..
does hat pump already have a stepper?
if so, it should have a matched driver, but more importantly, a matched power supply.
the Arduino stuff is beginners stuff as there are many ways and lots of help for this project.
if the pump has a stepper and power supply, post make and model, links etc.
if not, you need to make sure a stepper can run the pump.
flow. need what device you plan to use for the flow measurement.
if you have a transmitter, need data sheets. the output from the sensor is input to the arduino.
what accuracy is needed depends if the Arduino needs an amplifier board.
Robin2: The original pump has stepper motor and hence I am trying to use as much original equipment as possible. The pump in question was made by a Japanese company in 1980s and performed well for us; however, the control board is not working now, and the company has stopped support on it so I am being forced to be creative to make it work again.
dave in nj: The pump is basically a piston controlled by stepper motor; so for one revolution of the motor, the piston draws in a fixed amount of liquid; so in the original design I think there was no flow measurement device; all they did was monitor pressure, and change the motor current in such a way that the motor rpm stays the same even when the pressure changes. using a potentiometer, the speed (rpm) of the motor was changed and using some calibration curve, that determined the flow rate.
I am attaching the original service manual, page 13 shows the pulse pattern needed for the stepper motor, and page 39 shows the schematic of the original board whose function I am trying to replicate.
I have basic data acquisition knowledge using national instruments USB 6008 and labview but that's my extent of experience. I have played with arduino uno in switching LEDs off and on using both lab-view and IDE but I dont know enough to judge if I can produce the pulse pattern to control the stepper motor using arduino and combine that with some other driver board.
The original pump has a in built power supply, however, I'll buy a new power supply if that reduces the complexity of the current project; I have some PC power supply laying around, however idk if they would suffice or would I need something more high powered.
dave-in-nj:
application...application...application..
does hat pump already have a stepper?
if so, it should have a matched driver, but more importantly, a matched power supply.
Oh BTW, yeah the original pump had stepper motor with its driver and power supply but right now I am trying to look for a stepper driver which can accept DC power supply with wide range of operating specs and which can power my motor. We currently have multiple models of the same pump and all of them are roughly same wear and tear on them so it would be nice to come up with a robust solution with which I can retrofit other pumps as and when their boards/power supply etc fail.
bigdawg:
Robin2: The original pump has stepper motor
I have had a quick look at your PDF document. That circuit diagram is far beyond my pay grade.
Can you get access to the wires connected to the coils of the motor? How many of those wires are there? I would expect either 4 or 6. Can you measure the resistance of those coils and post the details?
What made you say it is a 5 amp motor?
My instinct is to buy a brand new stepper driver that can manage the necessary current and connect it directly to the motor coils.
Robin2:
I have had a quick look at your PDF document. That circuit diagram is far beyond my pay grade.
lol if someone at your knowledge level (basing on that excellent stepper motor basics thread) has difficulty with that schematic circuit diagram, than my situation seems hopeless.
Can you get access to the wires connected to the coils of the motor? How many of those wires are there? I would expect either 4 or 6. Can you measure the resistance of those coils and post the details?
let me try and get some values and I'll post them here.
What made you say it is a 5 amp motor?
well, there was a sticky note in the documentation binder which said its a 5 amp motor. I'll be honest with you, I've no idea what it means; whether it is the max current the motor draws or something along those lines.
My instinct is to buy a brand new stepper driver that can manage the necessary current and connect it directly to the motor coils.
I am willing to buy a new stepper motor driver; I just am not sure if a sparkfun board like https://www.sparkfun.com/products/12779 are available for the specs I may need. Either way, I need to be able to control any stepper motor driver board with an arduino which is already connected to a NI labview program so that the flow rate of the pump can be controlled as a part of the larger experiment.
Robin2:
Can you get access to the wires connected to the coils of the motor? How many of those wires are there? I would expect either 4 or 6.
I am attaching the relevant screenshot of the stepper motor information it shows four phase system. because there are four wires being shown in schematics going into the motor from transistors Q11, Q12, Q13, Q14 for A, A bar, B, B bar phase; I am assuming its 4 wires connected to the motor but let me open it up to take a look to be sure.
since you said your skills are on the Biochemist side and Arduino.... not so much, your choices are somewhat limited. the Geckodrive above will drive your steppers. they make a version with an on-board MCU so you can program that.
Looking at the instruction manual schematic, you have uni-polar motors with 20V. I am certain you can re-wire those for bi-polar and use the Gecko and get very good operation from them. in looking at the schematic, you have a couple, multi-pole switches, lots of op-amps and comparators.
certainly nothing that could not be done with different hardware.
One consideration is that analog control is often more accurate than digital (some will say heresy, but just stating facts)
Looking at the project, the motor driver is simple.
Deciding how to control, logic, and sensor inputs. That is the question.
the cost is not cheap. the Gecko drives will be pricey.
I would say that getting actual photos of the board and the stepper part number (photo of the label?) would allow us to offer voltage. certainly 24 volts at 5 amps is something you can get for a very reasonable costs.
bigdawg:
I am attaching the relevant screenshot of the stepper motor information it shows four phase system. because there are four wires being shown in schematics going into the motor from transistors Q11, Q12, Q13, Q14 for A, A bar, B, B bar phase; I am assuming its 4 wires connected to the motor but let me open it up to take a look to be sure.
dave-in-nj:
great manual. gecko drives. http://www.geckodrive.com/geckodrive-step-motor-drives/g203v.html
you have a great reference in that manual.
since you said your skills are on the Biochemist side and Arduino.... not so much, your choices are somewhat limited. the Geckodrive above will drive your steppers. they make a version with an on-board MCU so you can program that.
I'll go through the gecko drives and their step motor basics page; it seems like there is lot of information there so it will be pretty helpful. These drives do look on the expensive side, however, its worth it if I can find a way for it to "talk" with my current labview program which uses NI USB 6008 and arduino uno as a serial port controlling the old circuit which controlled the original motor.
Looking at the instruction manual schematic, you have uni-polar motors with 20V. I am certain you can re-wire those for bi-polar and use the Gecko and get very good operation from them. in looking at the schematic, you have a couple, multi-pole switches, lots of op-amps and comparators.
certainly nothing that could not be done with different hardware.
Thanks a lot for spotting that for me. I was under the impression that I have been dealing with a bipolar motor all along because I only saw four wires in the schematic instead of the two 20V ones. I have picked up a bit about electronics but most of my experience is dealing with using arduino for data aquisition into a Labview program so my apologies for not being proficient in providing right details.
One consideration is that analog control is often more accurate than digital (some will say heresy, but just stating facts)
Looking at the project, the motor driver is simple. Deciding how to control, logic, and sensor inputs. That is the question.
the cost is not cheap. the Gecko drives will be pricey.
I would say that getting actual photos of the board and the stepper part number (photo of the label?) would allow us to offer voltage. certainly 24 volts at 5 amps is something you can get for a very reasonable costs.
I totally agree with analog control being more accurate. Right now, my intrumental setup uses a digital to analog converter chip (NI USB-6008) so that everything can be controlled remotely using a computer; and my boss keeps telling me how we are losing accuracy because of DAC conversions.
About sensor inputs and stuff, I still havnt completely figured out how to go about doing it. I have a pressure transducer and a photocoupler on the motor shaft from which we can know the position of the motor; the bigger challenge is to use both these information in precisely controlling the motor.
I'll take some pics of the actual circuitry and motor so you guys can catch something I have missed.
A 6-wire unipolar motor can usually be operated as a bipolar motor by ignoring the wires that go to the centres of the coils.
The Gecko drives (Leadshine is another brand - may be cheaper) almost certainly take the same step and direction control signals from an Arduino as the cheaper and less powerful Sparkfun stepper driver.
If you really have a 5-amp motor the drives will be expensive. But if you have not had to spend money on the drive system for 25 years maybe there are some savings in the piggy-bank
I have posted some pics of the stepper motor and the controlling circuit from one of my pumps which is in working condition. The motor is from a company called japan servo motors co, and my co-worker tells me its undergone a name change and its now called NIDEC SERVO CORPORATION.
The service manual lists Vexta PH296-01S as alternate motor driven by the same control board although all the motors we have are from japan servo co.
dave in nj: you were absolutely right; it has 6 wires.
Robin2:
The Gecko drives (Leadshine is another brand - may be cheaper) almost certainly take the same step and direction control signals from an Arduino as the cheaper and less powerful Sparkfun stepper driver.
I'll contact leadshine since many of thier prices are not up online; they do look expensive but I guess I'll go with them once I know which board will meet this motor's specs.
If you really have a 5-amp motor the drives will be expensive. But if you have not had to spend money on the drive system for 25 years maybe there are some savings in the piggy-bank
...R
The issue is that I am now second guessing whether or not I do have a 5A motor on my hands. I have gone through the entire manual front to back and it provides no detail about amperage; the 5A figure comes from a sticky note from God knows who since all of these documentation files have been around for 20+ years. I would hope its not as high as 5A so a cheaper board might be used.
If I cant get the spec sheet from the original manufacturer, do you have any idea on how to get the operating current rating of this motor? in the document I am attaching it says "motor curent' as 3.6A for a high pressure scenerio and to be honest we dont expect the pump to be running at a higher rating than that.
the schematic lists amps as a correlation to pressure. this makes sense that the higher the pressure, the more power needed.
the snippet of your schematic that I cut, highlighted and posted (in post 10) shows the amps at pressure.
this is validated in your post of the motor.
Leadshine is a copy of the Gecko parts. Gecko chose to not sue for patent infringement, but rather developed a more advanced driver. the leadshine clone attempted to bypass the exact layout and got some things, mostly timing, wrong. but it has been over a decade since I saw those documents and comparisons of the boards. I would suspect that by now, they have figured out how to get it working. The boards from both companies used the same driver MOSFET and was capable for 10 amps. IIRC, there was a 99++ survivability under 7 amps, about 98+ under 8 amps, 96 or 97 at 9 amps and about 92 at 10 amps. since the survivability was so high under 7 amps, that was the rating chosen.
also at 80 volts and 7 amps, one was near the maximum one could get out of a stepper. more power means longer times to discharge the coil and performance was near it's limits. ( Failure was catastrophic and quite impressive, kill switches are recommended as something for the human to do while things go bad at the speed of light)
interestingly with the newer designs, the control of the motor took a step up. I think the normal applications never saw the difference, but the capability was there. think a Rolls Royce motor in a volkswagon beatle. both lasted forever, both were reliable, but one was just much more of a motor.
also the V suffex stands for vampire. one that cannot be killed. the idea was that almost all failures were from improper wiring and the internals for self protection were addressed to make the more bulletproof driver.
I have been using the Gecko for over 10 years. I think that either of these two would work in your application.
thanks for letting me stroll down memory lane.
I would offer that for one of your devices, using one of these would guarantee success.
you could play with others with the risk of not getting the motor to operate properly (skip steps under load)
geckodrives have a good white paper on power supplies and how a driver works.
8H=6FB-2
I think that is a gearbox on the stepper.
the photos are very interesting, the large green heat sinks on old transistors. not MOSFETS should not get hot enough to need that much aluminum.
the large capacitors are on the power supply board.
the output shaft (after the Lovejoy couling) has an opto interrupter on it. possibly a zero point. a place that one would stop the motor with the piston at bottom of the stroke. that would offer less possible stall opportunities on start and remove some stress on the pump when not used.
this is a guess, but it appears the disc in not slotted with hundreds of slots as if it were an encoder.
do you have photos of the face of the panel with the selector switchs ?
if you look at the panel face, you can get a feel for what will be needed for programming and inputs.
the manual discusses the pressure transmitter. the input is run into a comparator and this will turn the value into a sawtooth pulse. the speed of the pulse is part of the control of the stepper step rate.
it also discusses that the pump output will be a series of compression and suction. The transmitter will be outputting a VERY noisy signal that will need a lot of software to monitor the value. the waveform in the circuit is detailed as to show the stroke and the values on the 'quick return'
I have not really studied the manual, but it appears that they may only look at the output of the pressure while on the compression stroke to eliminate trying to filter the waveform of the output.
After the short study, the manual does show that the opto-interrupter, does show the bottom of the stroke.
the engineering around the problems this leaves me very impressed at the quality of the engineering from an age long past.
if you have a working unit, i would offer that a study of the values of the sensors would be in order to help you know what to expect so that when you attempt control. you know what to expect from the sensor output.
dave-in-nj:
the schematic lists amps as a correlation to pressure. this makes sense that the higher the pressure, the more power needed.
the snippet of your schematic that I cut, highlighted and posted (in post 10) shows the amps at pressure.
this is validated in your post of the motor.
Leadshine is a copy of the Gecko parts. Gecko chose to not sue for patent infringement, but rather developed a more advanced driver. the leadshine clone attempted to bypass the exact layout and got some things, mostly timing, wrong. but it has been over a decade since I saw those documents and comparisons of the boards. I would suspect that by now, they have figured out how to get it working. The boards from both companies used the same driver MOSFET and was capable for 10 amps. IIRC, there was a 99++ survivability under 7 amps, about 98+ under 8 amps, 96 or 97 at 9 amps and about 92 at 10 amps. since the survivability was so high under 7 amps, that was the rating chosen.
also at 80 volts and 7 amps, one was near the maximum one could get out of a stepper. more power means longer times to discharge the coil and performance was near it's limits. ( Failure was catastrophic and quite impressive, kill switches are recommended as something for the human to do while things go bad at the speed of light)
interestingly with the newer designs, the control of the motor took a step up. I think the normal applications never saw the difference, but the capability was there. think a Rolls Royce motor in a volkswagon beatle. both lasted forever, both were reliable, but one was just much more of a motor.
also the V suffex stands for vampire. one that cannot be killed. the idea was that almost all failures were from improper wiring and the internals for self protection were addressed to make the more bulletproof driver.
I have been using the Gecko for over 10 years. I think that either of these two would work in your application.
thanks for letting me stroll down memory lane.
I would offer that for one of your devices, using one of these would guarantee success.
you could play with others with the risk of not getting the motor to operate properly (skip steps under load)
Thanks for such a detailed post. I have learned more about stepper motor operation from this thread than by reading books on this subject. gecko's website has pretty good information about the subject too and I am going through the it right now. It is relatively pricey to get gecko boards for all my pumps; but from what you have told me, it looks like a well made driver, and since I am not a electronics expert, it will be perfect for me so I wouldn't have to worry about drivers and focus on logic and control system.
