Interfacing MO-CON501/Little blue TB6560 manual controller

After I fried my Chinese TB6560 board, I decided to move on to installing Gecko 251X drivers. I still had the little blue manual control that came with the TB6560 board that was very convenient for moving the axis manually. I am also using it to test the new drivers. My main problem has been interfacing the MO-CON501 pendant controller with a driver. With help from people on the "How I fixed my Chinese TB6560 board" thread, especially Doorknob, I am continuing to come up with a plan for this. Right now continuing to check out interface board per Doorknob's instructions. I already found that the step pulses coming out are only about 1v amplitude despite the 5v supply voltage. This wouldn't step the TB660 board and probably not the Geckos either.

Finally got a break on the problem interfacing this MO-CON501. There is someone on here that has thousands of posts credited whos signature line says something about persistence making obstacles dissappear. I'm starting to realize that I may be the only one out here with this controller. After reading the documentation for the tenth time I noticed a diagram that showed an additional menu that is mentioned nowhere in the written instuctions--it allows the setting of the output pulse parameters in the software when the pendant is connected to a computer. This is what you get when instructions are translated from the Chinese. Whoever translated this had no idea that this might actually be important I guess. I'll let you know what happens after I get this thing reprogrammed. I mean WT @*!?

That's good to hear.

As painful as it has been to get to this point, your insistent and persistent approach to troubleshooting this problem looks like it will finally pay off...

I finally had the time to get everything together for a test set up; Controller, drivers, and step motor. This is the controller that only had 1.6 v pulses coming out of it. I thought I could adjust the settings, but the documentation once again was misleading--no such menu actually existed in the controller. Didn't know what else to do but try it anyway--and it works. The attached pictures show the drivers mounted on my home made heat sinks and the test set up. The movie shows the pendant test using the x-axis output hooked up to my rotating table using manual move at a high setting and a low setting.
(Movie won't post--wrong format--I'll try again later)

Did you convert the rotary table for stepper drive?

If so, do you have any photos of the conversion that you could share?

Doorknob--this is a Sherline 4" rotary table that I bought "CNC ready" and put a stepper on later. So no--I didn't convert myself, just assembled it. I has worked really well--I used it hooked up to the TB6560 as a kind of lathe substitute, before I burnt it out. Turning the crank by stepper sure beats by hand.

I put the scope on the interface board with the driver board connected and without it. Just reference to the interface board ground with no driver board connected, the output wave is not as sharp and the amplitude is a little over 1 volt (the scale is 0.4 volt per division in this measurement.) With the interface board connected to the driver board and again referenced to the interface board ground, measured at the drivers step input terminal, the wave form is very sharp and the amplitude is 3.5 volts (the scale is 1.0 volt per division in this measurement.) I don't know what made the difference here but I'm glad I didn't give up based on the first measurements. Next step here is to mount everything, set up limit switches, e-stop, power switch, assess need for optoisolation, thermal and over current protection, and not break anything in the process. Almost there?;)

I would guess that your driver is open drain. You won't measure anything meaningful from it until it is connected to something that pulls up the signal.

Out of curiosity, as you slowly increase the step rate from zero up to the maximum, does the motor sound noisier (vibrate) at any speed or is it smooth all the way? My system have several frequencies where it vibrate substantially. It would be interesting to see how carefully engineered drives like the Geckos behave with common motors.

Thanks H500 for the info. I'll try the speed test you suggest and report back. Won't have time to start on it until next weekend though.

I ran the stepper from 1 rev/sec up to about 12 rev/sec and there is no change in vibration at any speed. The very slowest speed had minor vibration that can be smoothed out by adjusting a little pot on the board. It is smooth and quiet thru out the range. Also during this test I had the board and motor powered up for over an hour and the motor was only slightly warm to touch. I remember it getting much warming in a shorter period of time when I was experimenting with the TB6560 board.

Interesting. If the motor is barely warm, it might mean that you are not driving the rated current into them. Are you getting enough torque?

My motor vibrates at around .5 rps when driven at the rated current. Otherwise, it's smooth until it reach the midband resonance speed.

I don't know if they are making full torque because I have no way to measure that and haven't used this setup yet to cut anything. It certainly seems to have enough torque just from watching it change speed and direction. The rotary table will jump when the direction changes suddenly and its a heavy little chunk of metal. By the way I am trying to setup optoisolation for this now. I don't need another breakout board. I don't believe these geckos have built in optoisolation and neither does the interface board for the controller. If anyone has recommendation for simple and effective isolation in this application let me know. In the meantime I am stopping to research this and may end up building my own.

It should be easy enough to make your own using high speed optocouplers such as the hcpl-2531 and a few resistors.

I'v been looking at getting some hcpl2531--it seems to be recommended by several people--I don't want a problem with the wave forms like the TB6560 boards optoisolators and then have to fix that. The problem is that most of the suppliers of small quantities of these are in China and it will take two weeks or more to get here. When buy here in the US the prices are way higher. I just tried to buy on ebay from the "Little Diode" store and he wanted $9.95 apeice for them I wanted to buy ten so $95.00 for just the ICs?
I offered $2.50 ea and he came back with $8.50 ea. Should I accept? Or just bite the bullet and wait for some to come from China at $0.50--$1.00 each?

If I'm reading it correctly, Mouser shows various versions of that opto in stock, mostly for $2.50 or less in small quantities, and a price break at 25 units.

Mouser Electronics - Electronic Component Distributor hcpl2531

Future has them for $1.50. Next day shipping is about $10.

Search result for : hcpl-2531 - Future Electronics

Thanks to both of you for the info--but now I definately feel a little green, almost stupid really. Don't know why I thought my choice was only between ebay and China. I ordered from Future. They are not far from where I live.

Ok been studying info on opto isolation circuits all week and based on a schematic posted elsewhere (I think by Mariss from Gecko, but not sure because I've downloaded so much this last week), I came up with the attached schematic. Does this look like it will work? What about the warning that it completely defeats the idea of optoisolation if you don't have separate grounds? There is differing opinion here from an experienced moderator that insists that all grounds should be tied to the same point. Am I missing something here? I take this to mean that I need two or three completely separate power supplies unless the later is true and then maybe I shouldn't worry about optoisolation at all. What is the risk of going without it? How likely is a back EMI pulse? Would this happen if the stepper crashed under power against the mechanical limit of the talble or the Z axis hit the work at high speed?

I'm a bit puzzled by the PNP transistors on the output side, because they look like they're hooked up as NPN transistors.

I used MS Visio to draw this and the symbol defaults to PNP. I forgot to right click and change it to NPN which is what it is supposed to be. OK fixed it.