Disaster. I added some opto-isolators between the arduino and the sparkfun drivers to try

Disaster.

I added some opto-isolators between the arduino and the sparkfun drivers to try to separate the power supplies of the motors from the arduino.

However I was having trouble, and using my multi meter to check the connection for power. Originally I had the opto-isolators powered from the 5v coming off one of the drivers.
But that didn’t work. So I connected to the main 24v input.
I thought I’d solved the problems as the x and y acids moved fine. I was setting up to run a job, but moving the z axis failed. So I figured I’d check the village at the isolator for the z axis.

Stupid stupid, managed to slip and touch both the v+ and signal out lines of that isolator.

Spark. Fizz

This appears to have screwed up everything. :frowning:
All the drivers now pulse on and off. And signals sent to them get partially processed only when the power pulses on.

Not a good day

No idea yet if everything is really broken or just one thing. Wondering about upgrading all electronics and motors. But I really want it to be USB controlled and most kits seem to expect parallel ports which just send stupid to me. I don’t want to get a whole PC+monitor etc just to go parallel port. My laptop should be fine.

Wondering what the shapeko kits use. Living in the UK seems to be a significant disadvantage when it comes to buying CNC stuff

USB cannot support real time control so you need an on board microcomputer to do the motion with a USB setup. The imported TB6560 driver boards have built in optical isolation I believe so they are a plug in solution for you if you’re only running at 24 volts anyways. They’ll outperform what you have now if setup correctly.

Setting TB6560s up isn’t exactly trivial though.

@Paul_Frederick yeah I was thinking maybe a set of new drivers with built in opto-isolators. Searching for that part number brings up several different things from multi axis boards to single ones. Did you have a link to the specific version you were thinking of?

?.. is that case made of Styrofoam? Bad juju for electronics.

Sucks about your setup blowing up. Looking to upgrade all my stuff as well.

@Daniel_Would
I made my own drives and only know about the manufactured ones through what I’ve read on the net, so I’ve no first hand experience I’m afraid. Although my experience does give me some insight into these manufactured boards. I know a lot more about what is wrong with them than even the people who originally designed them I think. That is not to say that they’re all bad, but there are parts that could stand for some improvements.

The absolute worst part about the manufactured boards seems to be the documentation that comes with them. I’ve read some of that online. Knowing what I know, and comparing it to what they say there are gaps in places. My advise there is to get Toshiba’s original documentation and learn that yourself. No one knows that drive IC better than Toshiba does.

The biggest issue with any manufactured TB6560 drive board is with the current sense resistors and how adjusting the output current is handled. The sad fact is there is no easy way to adjust that current because of how Toshiba designed the chip. They designed that chip to be put into custom builds. Something like a VCR where the designer knew exactly what motor it was going to be driving etc. When that is the case everything is rosy, unfortunately with hobby CNC things get muddy.

These drive board manufacturers have taken it upon themselves to outfit their boards with sense resistors then tell their customers to adjust current using the digital input lines. That quite frankly is a crock of crap they’re peddling there. It works but it is less than ideal. Toshiba put those lines on the IC for software motor control, not to adjust the overall current.

I annotated this image in red to highlight the current sense resistors

Note the manufacturers annotation in the neighborhood of the fan on the heat sink “Operating current sub - setting” That is a fabrication on their part. Yes it will adjust output current in crude quarter steps but it isn’t what any CNC builder really wants. Better to leave that at 100% and just have the right sense resistors installed. Toshiba’s documentation explains how to calculate the correct value for your motors.

But for what they are the multi axis imported boards are pretty impressive never the less. It just takes understanding what they really are, as opposed to what the manufacturers are saying in order to realize their full potential. The key is in due diligence on the part of the integrator. I’ve read many horror stories online where people didn’t fully appreciate their responsibility in the equation.

They could make a Fox show about all of this, When Good Drive ICs Go Bad. The moral of course is the ICs are good, the boards are even pretty good, it is people who are bad. The people making those boards, and lots of the people using them too. What I’m saying is if you do your part you can salvage this situation to your benefit.

And yeah, looking at that board it is fully optically isolated too. You can see the 4 leg DIP ICs on the left hand side of the board. Those are all single channel optocouplers. Although how the board handles crosstalk between the drives on the board itself I’ve no idea. I imagine it does though. They’ve cunningly placed electrolytic capacitors around it I suppose by the looks of things. Even on the output side. Interesting.

@Peter_Fouche1 I’m very tempted by something like that. But to @Paul_Frederick point, increasingly I realise that I need to understand the specifics of what a board is doing to truly build a reliable system.

@Peter_Fouche1
It looks like you went with a vendor locked in solution to me. If you wanted some smoking hot hardware you should have gotten one of these http://www.mesanet.com/fpgacardinfo.html

So I’m pretty sure my problem is that I’ve fried one of the easy driver boards. I just tried powering from a 9v battery and the magic black smoke was coming from the one I accidentally zapped.

So I probably should buy a decent set of drivers, either off the shelf or built to one of @Paul_Frederick schematics.
Or
I can buy another £10 easy driver to drop in my existing setup and try to fix the wiring on the one opto-isolator that want working.

I think one day soon I’m going to build a new machine. With significantly upgraded electronics and mechanics. But I think that will have to wait until after I move house. For now cheap wins the day. But I will try to accept I am the architect of my own frustrations…

@Daniel_Would
I’ll keep it short, and sweet. Neatness counts. Your build went off the rails at some point too. It looks like it started out neat enough. Now, not so much. I think you have to go back to square one and start over.

Look at this first build as practice. I can just about guarantee you that your second shot will be 1000 times better. Heck, it might even work. If not the third time is often a charm they say.

You haven’t failed until you’ve given up.