Stepper Motor Question

Please explain more about the set-up. Are there 4 motors? Does each have its own drive circuit?

Any chance you can post a block diagram?

I do not understand where these transistors reside in your circuitry. To speed up our game of 20 questions I will assume that you've made a common mistake and are not using a stepping motor driver board but are just applying the motor coil directly to your power supply with the transistors you are toasting sink current directly back down to your power supply return.

Most people grasp the idea that they must sequence the four circuits per motor to obtain motor rotation. The repeated sequence for full step motion in one direction is A, B, A*, B*. To get rotation in the opposite direction, one just applies the sequence in reverse order. The multiple problems with this circuit approach hinge on the fact that people forget that you are not driving a resistive load, but an inductive load that also includes a back EMF once the motor is moving. (I'll first ignore the back EMF.) In turning ON each individual inductive load, the time constant for the current to flow through this inductor is Τ=L/R. With many motors having an inductance of decades of milli-henries combined with a series resistance in the range of single digit ohms or less, many times this apparently efficient circuit design means that three times Τ occurs in tenths of a second. Remember that three times the time constant is when the current reaches about 95% of the steady state current and that the current establishes the torque the motor produces. This is the amount of time it takes to make one of the presumably 200 steps required to make one revolution of the shaft of your motor. So if you try to step faster than two times the time constant most of power dissipation will actually happen in heating the power transistor and not applying torque to the motor shaft.

Now there are many linear and non-linear methods used to increase this process. I've made an assumption on what the circuit is that the OP is having trouble with. But my real point is that there is a very good reason that people make separate drivers for stepping motors. The dynamics that happens here are not as simple as it would often seem.

Its a simple Step Genie board. There are 4 Mosfets that are driven by the control chip. We designed a simple small fan cooling system and when all motors(4) are driven in the same direction the transistors/motors remain cool. When 2 of the motors are reversed driven (yes we had it working) they remain cool, but on idle (holding torque) the transistors heat up.

In honesty i am an ME, not an EE, trying to get this system to work. My programmer is going to change the instruction set to no holding and eliminate the idle step.

Thanks

Ron

I knew it. I make an assumption because there's not enough information to help. Go through a lengthy analysis with my assumption and viola my assumption is wrong and therefore my analysis is just entertaining.

Many times with a stepping motor system there is no need for a holding torque at all. This is because very often the static frictions will easily hold everything right where they were moved last. More often than not a small amount of holding torque is desired so that accidental contacts with moving parts will be resisted but not solidly held in place. But occasionally that several kilogram mass will slide to the bottom of the stage travel if the motor holding torque is reduced even slightly. In these cases I prefer to include some type of a spring loaded brake that will at the very least share the load with the motor but this requires more effort in the programming. Now the power supply feeding the driver and motor combination is often providing the same amount of energy in holding something in place and moving something around. Since holding something in place means not doing any work on that object but this energy from the supply has to be going somewhere, the driver transistors get hot. So if in your software settings you tell the driver to work as hard as possible to keep things in place, then that requested energy must go somewhere.

I more-or-less assumed this - unfortunately the Step Genie doesn't appear to have any means of reducing the current to a "hold" value, even if it was mechanically an option.

The thing that is puzzling is that (if I understand correctly) the temperature rise is only troublesome when 2 of the motors are connected for reverse operation. Makes no sense at all.

To the OP - are you connecting all the motors in parallel? How many series resistors are you using?

unfortunately the Step Genie doesn't appear to have any means of reducing the current to a "hold" value

Well the motors wont turn when the supply V is on( i had to crawl inside the pipe and drag the PIG out). I am assuming (we all know what that means) that the problem lies in the program.

Makes no sense at all.

I agree. My general application is reversable DC gear motors. However we had to get our "feet wet" for another project using steppers that project being a portable ultrasonic plate scanner with C scan reporting.

Ron

Not by itself because the Step Genie only produces the step sequence. (I looked up the board with Google.) First, I'm stunned that anyone would design a board with four TO-220 package power transistors and not provide room for heat sinks.

Back to how to adding a hold current setting circuitry. The two power resistors on the right hand side set the current in the motor. If you add a two pole relay that adds another pair of resistors in parallel to these two resistors then you can have a bit I/O (for each or all motors) of your controller select the fixed and running current. You could also switch between these two modes automatically. I do not have the time to draw up a full schematic now but if you fabricate a retriggerable one shot circuit (a 555 timer chip should do it) to trigger on the step input signal coming into the Step Genie then this relay would toggle every time a pulse came into the Step Genie. Search for a missing pulse circuit diagram. When the pulse is missing you'll be in hold mode, when the pulses are present you'll be in run mode.

Re: heatsinks - to be fair to the Step Genie guys, it is only a developement ("Demo") board. It would be fine with a "sensible" motor supply and e.g. a 16 or 17 frame size motor for doing lashups on the bench to get a feel for driving steppers.

That said, not a bad idea with the relay for switching the resistors. A solid state solution would be better, but (agreed) it would get messy with e.g. P channel MOSFETs "up the top".

Can't help feeling that driving several motors in parallel (which is what I assume is happening - not having had a reply from my query) is asking for trouble.

Re: heatsinks - to be fair to the Step Genie guys, it is only a developement ("Demo") board. It would be fine with a "sensible" motor supply and e.g. a 16 or 17 frame size motor for doing lashups on the bench to get a feel for driving steppers.

That was the original idea. It was handy and the tech wanted to drive 4 # 23 motors. He had to go to a much larger PS and i designed a cooler for the transistors and it worked well. I had the tech go to 2 power resistors yesterday after changing the program and we did 10 runs down a 40' pipe and everything worked well except for a slight vibration that i worked down to a design error on the frame.

Holding torque is not necessary on this application as the unit dosent have enough mass inertia to change the location measurement,when stopped, and the PIG automatically reverses at the end of travel.

Thanks

Ron

Glad to hear it's working ok.

The design that DID NOT work. As this was a proto we wanted to go with steppers for the precision that is not produced with gear motors. My first (above) design i had gas spring loaded 4 of the NEMA 23 motors at every 90 degrees. With a total loading > 30 lbs the motors had trouble indexing the PIG down the pipe. We were direct driving 4" diameter wheels and clearance on the pipe ID was a problem with larger steppers or gear motors.

The design was changed and now i believe we are on the right track.

Thanks for the help

Ron

Don't forget you can always bung gearboxes on the 23 frame motors if you start running low on torque - could have a problem with speed tho' (for reasons Redfred said), using L/R drive.

Don't forget you can always bung gearboxes on the 23 frame motors if you start running low on torque - could have a problem with speed tho' (for reasons Redfred said), using L/R drive

Again a clearance problem. I am attempting to apply KISS. and also keep the cost down.

1. Make it light.

2. Make it simple.

3. Make it accurate.

4. Make it reliable.

Most of the PIG designs are developed by someone sitting behind a desk and they havent seen the enviorment that these units are subjected to.

My boss approches me about 4 months ago and says " design a man portable PIG that can index down a pipe and measure changes in the ID". This was after a major problem in the industry where after Hydro test it was thought that the pipe expanded in given sections. My boss believes that it left the pipe mill in an already expanded condition. The cost to pipe suppliers ran to many millions of $. So i have to get this PIG working.

Again thanks for the help.

Ron

I know it's not down to you, but don't you just hate it when someone says:

"The cost ... ran to many millions of $ ..."

... then just as you're going out the door they say:

"Oh ... and keep the cost (of the solution) down"

... and they leave you with a budget that would just about stretch to a couple of pints and a pizza.

Glad we could help.

There's one additional thing I do want to point out though. I thought you mentioned somewhere that you are considering using DC servo motors in this design. Be careful if you switch to a DC servo motor design. Many of these motors contain brush commutation which produces some very significant sparking inside the motor. Sparks and the environment of a PIG can produce some very undesirable results. I don't know of any stepping motors that have this as a potential problem.

You have described the problem exactly, the holding current is too high. May I suggest that you change to one of the stepper motor driver chips from Allegro and enable the feature that turns off drive to the motor if the motor is not to move.

I expect you have direct drive (probably a bad choice) and need the holding current to keep it in one place on a slope....

A better method would be to gear each motor down say 10:1, then the holding torque with no power will be enough to hold the vehicle on the slope......play around with the values till you get it right.

By the way, Unipolar motors can often be driven better in Bipolar mode (or buy Bipolar motors!), which gives more torque, or you can use a smaller motor and get the same torque.

Don't forget, you are also wasting power when standing still as well, especially bad if battery driven.....

Some chips allow the hold current to be reduced instead of just turning it off, some allow it to be variable.....

Just in case you don't know about it a great reference is:

http://www.cs.uiowa.edu/~jones/step/ Jones on Stepper Motors.

Yes, a very good reference.

The driver has to be matched to the motor sometimes u will use H deiver

daveross100atgmaildatcom

A usual type H driver is for DC motors, usually commutated ones, sometimes with an encoder for positioning.

They are not for stepper motors.....the question was for steppers......

Steppers have in their construction a distinct similarity to PM AC induction motors (many will never understand that comment!!), that only becomes really apparent when you use one as a generator (the PM ones at any rate!)......

Added two motors on the right hand side. Went to solid adjustment struts. Changed the program. Now Rover PIG works.

Thanks

Ron

To be fair to guest Dave Ross, you can use H bridges for a (bipolar) stepper, but you need 2 of them, as in the L298 dual H-bridge driver.

That was not mentioned.......therefore I ignored it.

If he meant two H switches, then he should say "2 H switches"........50% of an answer, is not an answer....its also not a pass mark in the Uni or school!!!

It also might mislead other people......

As an update. Rover is in use and measures pipe length,ID,ovality and bow in one pass.

Ron