I have looked at the MD22 manual, hook up diagrams, and the schematics. It looks like as straight forward a design as all the rest.
Can you post an image of your wiring. I do not believe the problem has anything to do with the received pulses, but with the logic on the motor controller or microcontroller possibly resetting due to a power / wiring issue. A picture is worth 1000 words. This forum will even host it for you to make it as easy as possible. Can you do this?
Hmm, i didn’t think of the SSC possibly resetting yet. I’ve seen it do that months ago when i still had the VL=VS and the battery went empty, didn’t notice that behaviour this time. I’ll hook them up to a more juicy power-source and will let you know.
E: No luck there, SSC and MD both on the 5V from a PC PSU. I think we can rule out power-drops/resetting…
And for a pic, that would be just a snapshot in time, i may have rewired it all by tommorow.
This is how i got it wired now, excuse my drawing skills…
Ok a couple things wrong here… When connecting the ground of TTL digital control and high current motor drive systems it’s best to pick a single common point, and rout all grounds to that one point. This prevents ground loops. However that’s probably not causing the problems you are seeing. The MD22 requires 5vdc for the logic supply. The SSC-32 logic input is actually feeding a LDO regulator. It requires about 1 volt above the desired output voltage, in this case 6vdc. For a simple test power the SSC-32 VL input with a 9vdc battery. 
Alternately it may be possible to feed the 5vdc into the +/- pins by the ABCD inputs. I’ve never done this so be careful.
How many forums did you say you posted to? lol 
I hope this information will renew your faith in hobby electronics / robotics. 8)
In other words, its kind of hard to try to regulate incoming 5v through a 5v regulator and expect a steady 5v out. There’s likely to be some voltage drop through the regulator so 6v should help give you that 5v you need.
Ok thanks, didn’t have a 9V battery so i tried VL=VS like this:
VS1 = 6.15V and the MD22 gets 4.99V, still no normal behaviour (although i’ve seen worse).
This should make the motors run right? (in servo-mode with timeout, on-on-off-off)
Nothing happens…
I don’t know anything about that software. If it’s not too much trouble could you download and install our simple terminal program LynxTerm?
lynxmotion.com/Product.aspx? … egoryID=15
Then you just type #0 P1500 #1 P1500 cr
Or you can unplug the motor controller and plug in a couple servos. They should be at center of rotation when sent 1500uS pulses. It’d be a good reality check that the SSC-32 is working properly.
I’m not sure what the current draw is on the MD22, but the SSC-32’s regulator should be able to handle it. You should be able to draw 100mA extra without issue.
Also check your baud rates…
Well lol, this is just a little test-app i made apart from my robot-app. I’ve been working with 2 servos+cams for months, tracking objects etc.etc. never had an issue with those. When i check box 0 or 1 or both then the appropriate servo moves, just as it should. It’s not the software causing trouble.
Putting a servo on channel 2 or 3 works just fine.
The SSC-32 has a green LED by the DB9. It’s not a power light but a status indicator. It’s job is to light up when power is applied, then when it receives a valid command it will go out and then blink when receiving data. If it’s resetting you can notice it because the LED will be on steady. Do you notice anything like this happening?
This is why I suggested using lynxterm. It will send a single command and you can see if the SSC-32 is resetting.
The image of the wiring is pretty, but it does not allow me to see if there are “problems” with the wiring. If the wire you are using is like 30 gage it will never be able to carry the required current and all sorts of resetting will be observed. This is why I asked for a picture instead of a drawing.
The original problem, 5vdc into a 5vdc regulator should have been the only problem. But it looks like there is another problem going on at the same time.
Maybe a video is the next step here…
Then there must some other “assumption” that must be incorrect. Generally it is either a hardware problem or a software problem. Hardware wise, things uaually work they they are wired.
It’s possible the motor driver has become defective, but who knows.
I’d start with a photo of the setup before moving forward.
Thanks, pics i could provide tommorow i think. All wires that need to handle some current are either wires from a PC-PSU or supplypower-wires, from an old tv for example. Sure they can handle a bunch of amps. The signal don’t need these kind or should i just try that? Tried some shielded wires already though, core aint very thick though.
Just in case, is there a way to make the SSC32 output-range (analog measured) from 0V-5V?
The frustrating thing for me is I know I have one of those stupid MD22’s around here somewhere. Haven’t seen it since the move. 
I would be happy to test it here if I could find the damn thing. 
Frustrating for you… Got a $1000 robot which can’t drive because of a SSC32-MD22 issue 
Ok I might have a clue as to what is going on. The SSC-32 generates the pulses for each bank of 8 servos at the same time. So all channels in a bank 0-7, 8-15, 16-23, 24-31 go high at the same time. Try putting one channel on pin 0 and one channel on pin 8. This will ensure the two pulses are not going positive at the same time. It’s a long shot but worth a try. You can also put the second channels on pin 16 or 24.
BTW There is a lot that you could do to help troubleshoot this. Have you installed lynxterm so you can try some things for me?
Give that man a sigar!!! Woot!!! It works :mrgreen: !! Finally, after weeks… used channel 16 & 24 and tried servo-mode and servo-mode with timeout, both work 
Thanks!
(maybe put this in a sticky or something for future troubleshooting)
Whew! Wipes sweat from brow… 8)
I’m glad you are now able to progress with your project. It just dawned on me over the weekend what it was. No discussion of faith being restored. I don’t want to jinx it. lol
I don’t know why Devantec tied down the pulse requirements so rigidly. It’s basically programmed to ignore the second pulse if it either occurs too late or too early. Even newer 2.4ghz radio receivers don’t necessarily provide the pulses in sequential format anymore. That’s simply a bad design…
