Need Help!- adding a chopper type current limiter

[gandalf1]

I have a very basic system, 3 separate controller cards opto isolated from the PC which works well enough using dropper resistors. The motors are Bi-polar with parallel coils and the rest of the machine is more or less conventional but no limit switching, homing or even an E-stop as it is quicker to stop the machine by grabbing a hand wheel on the end of the appropriate feed screw. I would like to try a 'chopper' circuit to limit the motor current to 1 amp per coil rather than the dropper resistors but without having to buy or build completely new circuits so would like to ask those with far more knowledge than myself if it is possible to add such circuits to the output terminals of my control cards. Maybe it cannot be done this way but if anyone can help me and lead me to a suitable circuit (board layout and components) I would be most grateful.
On the other hand if it cannot be done simply then that information is almost as useful.
thanks in advance.
John.

[irving2008]

John,

I have a feeling this could be done, but I'm not clear on your setup... a schematic would help...

[gandalf1]

The boards are Quasar 3158 and the full desription can be seen on
http://www.quasarelectronics.com/kit.../3000/3158.pdf
i tried a picture but could not work out how to attach so give the web address instead. When I started they did not sell a chopped board as they do now but as they charge more for a kit than ARC Eurotrade do for a finished driver with better specification I might have to save the pension until I can justify the £90 for the three so you can see my interest in something I can add on.
Hope the address helps.
John.

[keebler303]

Someone on the board has done it, I remember seeing it. Maybe try searching?

I'll see if I can find it.

Edit:

Check out the chopper adapter here:

http://pminmo.com/3axis/3axis.htm

The designer goes by tachus42 here on the zone.

Matt

[irving2008]

John,

Yes this could be done. I don't know how competent you are with electronics etc. although I get the impression you built the Quasar boards from a kit?

This link gives a good overview of the approach http://www.cs.uiowa.edu/~jones/step/current.html, fig 4.10 being the best option

You might need to use the spare gates in the 4030 package and a daughter board with the comparator on it. The comparator has to inhibit the upper driver transistor in each pair (the A transistor). Unfortunately the design of the Quasar board has the upper driver transistor turned on passively by the 2.2k resistors e.g. R7 and forcibly turned off by the inverting transistor e.g. Q1 so your chopper would have to do the same. One approach I have shown in the schematic below... but you'd need to be careful how the ground wiring for the lower pair and the sense resistor were arranged to avoid any nasty ground currents etc. Apologies for the crude update of the Quasar diagram but its only to give an idea... there are lots of other ways to approach this. The idea here is that when the sense voltage on the comparator gets higher than Vref the comparator turns on, forcing the transistors Q1 and Q2 on and turning off Q1A and Q2A. The energy in the coil is returned to the supply through the new diodes (I'm assuming that Q1A etc dont have built in diodes - they may well do). The feedback resistor between the outut of the comparator and its positive input provides a level of hysteresis that allows the energy in the coil to decay until the sense voltage drops well below the threshold. Adjusting Vref adjusts the current sense point. I've not shown any values as these need to be worked out, but assuming a 1v sense the sense resistor is 1ohm per amp (and wattage rated accordingly) and Vref would be 1 volt. If the comparator input resistors were 10k the feedback resistor would be maybe 1M and the output resistor 4k7, but some experimentation may be required.

[irving2008]

Originally Posted by keebler303 Someone on the board has done it, I remember seeing it. Maybe try searching? I'll see if I can find it. Edit: Check out the chopper adapter here: http://pminmo.com/3axis/3axis.htm The designer goes by tachus42 here on the zone. Matt

Thats another option, but its a lot more complex (admittedly doesnt require mods to the Quasar board) and a bit of a sledgehammer approach. I'm not keen on the idea of chopping the main supply line to the board, means you have high current transients in the leads between the chopper board and the Quasar board. Forcing off the upper transistors in the H-pair is a softer approach IMHO.

[gandalf1]

Whilst I am no good at circuit design (nor copper layout) and electronic theory either I can usually follow the schematic with some degree of success and I have a spare board that I can 'unmodify' to return to the original configuration. Daughter boards are no problem if I can use perforated strip board nor are etched ones if I have the copper pattern to transfer to blank board. I do in fact already have extra boards fitted to the Dir pins that are little inverter circuits since the MasterCNC software timing was not suitable for the Quasar boards and odd shaped circles were the result.
Now I have to decide on whether to attempt this add on, save for months for commercial drivers or to go Uni-polar with new circuits so I suppose pricing up componentsd from Maplin is a good a starting place as any.
Thanks for your help and thoughts.
John (Not far from Peterborough)

[irving2008]

Originally Posted by gandalf1 Whilst I am no good at circuit design (nor copper layout) and electronic theory either I can usually follow the schematic with some degree of success and I have a spare board that I can 'unmodify' to return to the original configuration. Daughter boards are no problem if I can use perforated strip board nor are etched ones if I have the copper pattern to transfer to blank board. I do in fact already have extra boards fitted to the Dir pins that are little inverter circuits since the MasterCNC software timing was not suitable for the Quasar boards and odd shaped circles were the result. Now I have to decide on whether to attempt this add on, save for months for commercial drivers or to go Uni-polar with new circuits so I suppose pricing up componentsd from Maplin is a good a starting place as any. Thanks for your help and thoughts. John (Not far from Peterborough)

John,
If you decide to go down the 'build it yourself add-on' I'm happy to help with design as far as I can. I'm NW London based. I am just starting the design of my own stepper controller boards (I won't buy ready made unless I have to ) which will probably be based around the Allegro chips although I have 'knocked up' a basic H-controller to allow me to measure the stepper torque and to check alignments etc.

The solution I suggested is about the simplest you can get away with. There are refinements but if all you want to do is get rid of the ballast resistors then that will work. How big are the ballasts, supply volts and what steppers are you using?

regards,
Irving...

[gandalf1]

Hello again Irving (christian name?)
As you worked out all I wanted was to get rid of the resistoprs hopefully by simply adding a current sensing device to the motor leads anyway what I use at present are 3 off Slo-Syn MA61FS-80020E motors 60 oz hold. Five Volts 1 Amp bought from Smartdrive(20 mins drive away) some years ago (box of 5 for £15). The power supply is from Maplin 13.8 Volt 10 Amp (just about big enough) and the resistors are 8R2 25 Watt . The coils are set up as Bi-polar parallel with one resistor on each motor lead from the Quasar board. It works but does resonate although that in itself for me is no problem it being outside my normal stepping rate. I thought the motors did get excessively warm so fitted a PC power supply fan on each motor and now thre do not even seem to get warm. The resistors are not near the cards themselves but fitted to the aluminium shroud that surronds each motor to duct the air from the fan alongside the motor casing and that of course helps keep the resistors cool as well.
Cannot think of much eslse other than 700 steps/sec, motors geared down by toothed belts 16:24 to give 0.01 mm/step using 12 mm x 3 mm trapeziodal screws.
Shopping time now.
John.
PS Used to work at W.G.C. so not far from where you live but that was in another existence and 100 miles a day driving.