I will be driving a unipolar stepper motor with an old 68HC11. I'm thinking about using this
formatting link
circuit control the motor so the microcontroller can focus mostly on sensor input. I've been told I need to isolate the microcontroller from the stepper coils via optoisolators. I have three, possibly three questions about this:
I'm assuming this means I will need two power supplies, one for the microcontroller/sensors and one for the steppers?
The optoisolators I've looked at don't mention a built-in resistor in the schematics, do I need to provide one
Would I put the optoisolators between the microcontroller and the control circuit (IE on the clock/step and reverse lines) or would I need to also isolate the controller circuit (IE place the opto- isolators between the JK Flip-flops and the power transistors?)
If I need to isolate the controller circuit as well, would the
4027 be able to sink the (10ma if I'm reading the 4Nxx spec right) current from the opto-isolator? I don't see info on maximum power dissipation on the 4027's spec sheet
formatting link
4000Series/CD4027.shtml) though I admit I'm no EE.
1) You'll need two power supplies anyway, unless your 68HC11 operates on 12VDC. Inductive loads like motor coils create voltage spikes and sags on supply lines, which can cause uC problems. Optoisolating the uC from the inductive load is probably a pretty good idea.
2) Optoisolators usually have an LED on one side, and a phototransistor on the other that picks up the LED's light and turns on. You do need a current limiting resistor for the LED, otherwise you'll stress out the uC output port pin.
3) You can put as many or as few optoisolators wherever you please. Usually it's done right at the STEP and DIRECTION signal, because that means you'll only need two optoisolators.
4) Probably a bad idea to drive an optodiode with a 4000-series CMOS output -- it just can't source/sink enough current. Again, best to drive the optoLED from your 68HC11 output.
Now for the rest of the story, as Paul Harvey would say. First off, the 4N26 is a miserable choice for logic optoisolation. Its current transfer ratio (the relationship between how much current is going through the LED and the current you can coax out of the phototransistor on the other side) is a ruinous 20% -- you have to pump five times as much current through the LED as you get at the output.
Look -- use the KISS principle (keep it safe and simple). The H11L1 is an optoisolator with built-in schmitt trigger logic output. It's been around for a long time, and it's relatively inexpensive. Drive the LED with only 3mA (for a 5V supply on the 68HC11, use a 1K series resistor from the port output pin to the LED). hook up the H11L1 pin 6 to the isolated +12V, pin 5 to the isolated power supply common, and attach a 10K pullup resistor from pin 4 to pin 6. You can then read a guaranteed logic level for your step and dir driver control signals.
The other issue is that you've probably got the wrong transistors there. It looks like you're asking the 4027 outputs to drive 10mA or so. That's not realistic. Not only that, but the current gain of the TIP31 transistors isn't going to be able to drive any but the smallest stepper motors, even with 10mA of base drive current. Do yourself a favor. Replace R1-4 with 4.7K resistors, and use an NPN darlington transistor like the TIP121 for Q1-4. Your transistors will be glad you did.
If you use one supply, there is not much reason to isolate a signal and then reconnect it to the same supply.
Probably. Most isolators are an LED and a photo transistor of some kind. You have to add the components it takes to drive the LED with a reasonable (for it) current from your signal source and to produce an output signal from the transistor.
I think you should isolate the step and forward/reverse signals, if anything. But with correct supply connections, you probably don't need the isolation, at all. Use two supplies, one for the logic and one for the motor, and connect them at one point... near the emitters of Q1-4. The CMOS logic shown could be powered from the 5 volt processor supply, if you can get enough base drive through R1-4. You might need to add a current boost stage if the logic chips cannot supply enough base current to saturate transistors.
Doubtful. That would make the chips pretty warm, even if they could. Add emitter follower outputs and they can drive lots more current.
I'm curious about the dual power supply situation since this will be powered via batteries. Can I just plop a 7805 in front of the microcontroller and optoisolate as discussed, or do I really need a second set of batteries?
Also, do you know of a multi-package for the H11L1 so I can keep the chip count to a minimum? I was eying the ILQ74, but it of course doesn't have the built in Schmitt triggers.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.