Hand slipped busted something! :-(

[maclean]

Hi Everyone.

After finishing my VCA I got round to testing it by connecting a lead to R3 on the power supply and touching south arm of R146 to trigger a note. It was working (and really sounding like tb303).

However, my hand slipped and all I heard was the last triggered note ring on and then it would not trigger again... If I touch the north side(evectivley bypassing R146) I get a long WAAAAAARRRRRPP noise, but not the short plucky sound it should be.

Really don't know what the hell I done to break it, maybe the clip touched something and busted something! I really don't know. Anyone got any pointers for debugging this section?

After this broke I measured the voltage of R3 which i thought was a bit high measuring at 20+ volts, I was thinking it should be 5volts or something like that?

One other piece of information that may or may not be significant, is my AC/AC adaptor had a blue led when turned on... Since this test that blue light is no longer working, which is why I think ive touched something and busted something.. I tested the adaptor though and it is still putting out 9.5VAC.

Man its frustrating when this happens, but it really does get the cogs turning when trying to figure out what the hell is wrong!

[MMM]

I can't say for sure what exactly is wrong here (yet), but you may recall your previous problem.

"the voltage at R3" isn't exactly clear since R3 has two sides. And yes - one side is about 25 volts, the other one about 5.3 volts. Please check if this is the case; and also check the voltages coming from the IOboard.

If you've used the "wrong" end of R3 (read: the 20+ volts) to apply the voltage to R146 it could really have fried some parts. I'll have to check this.

[maclean]

Thanks for your quick response MMM.

One thing I can say is that the lead from my IO board seems okay...

When powered up (after the incident lol) it is still measuring 12 volts on one side and 5.33 on the other.

When applying my control voltage to R90 as usual the tone is still being generated fine and can be heard at the resonanace pot... I can only assume that If I have damaged it, it must be somewhere in the envelope or VCA.

Something else that I tried which was possibly a bad idea but at this point thought it was broken anyway...

If I set it up for testing as usual, and instead of using R3 use another lead from my 3.3control voltage to hit the north side of R143, this makes a very tb303 sounding note strike(however the pitch drops slightly due to the control voltage being used on R90 and to strike R146). But if I take this and strike the south side of R146 it still produces nothing.

[maclean]

Further development - After going through each component one by one I discovered I had C55 wrong way round and that was bulging so I'm positive its bust. I've replaced it and ordered another for when I need another later - but its still not fixed the issue.

I have also re soldered R3 with the 5.3 volt side up. Still not getting a thing when touching south side of r146 and now a very faint trigger sound when hitting north side.

So I guess between me touching r146 with a 20 volt trigger and having c55 the wrong way round I have likely fried some parts in my envelope and/or VCA, question is, is there an easy way to determine which components will have been fried by looking at the schematics?

Should I be looking at ordering complete replacement envelope and vca?

[MMM]

Should I be looking at ordering complete replacement envelope and vca?

Not yet. Hang on.

I do seem to recall you don't have an oscilloscope, just a multimeter? Debugging this part of course is easier with an oscilloscope since we're dealing with varying signals (envelope, trigger...). I'll try to think of something.

[maclean]

yeh I don't have an oscilliscope but do have a multi meter, I appreciate your help if you come up with anything.

The scopes looked a bit pricey, and I don't know how much more electronic things I'm going to be doing after my xoxbox hence my hesitance to buy one... But I'm quite determined to finish this xox one way or another.

I'm going to take a break from it tonight but will keep my eyes on the forum encase you think of anything, and again really appreciate any help you can come up with!

Thanks

[orange_hand]

Hi,

a pre owned scope is about 100 EUR's / 130 USD (e.g. Hameg 203 Series). If you want to continue building those kind of devices, you should consider to buy one.

Apart from that I would use the voltage overview chart posted here:

viewtopic.php?f=12&t=12453

I guess this will help you to identify the faulty component / area on your board.

Cheers
orange

[MMM]

So we'll take this one step at a time. We're dealing here with both (fixed) DC voltages and changing voltages (like the envelopes). Let's debug the whole thing from the start and see where things get lost.

Double/triplecheck everything. Possibly use test clips with hooks and don't slip.

WITHOUT applying anything to R146 measure the voltage of the collector of Q37 (this is the middle pin). This should be around 12V. Is that correct?

Now apply the 5.33v to the "south" side of R146 (the side towards the processor). Does this change the voltage on the middle pin of Q37?

[maclean]

Hi Orange hand, I think I will look into getting a second hand one, I'm not quite happy throwing £200+ on one just for this project, I guess i would like to build some other synth type projects but I actually think that even if I had a scope I'd still be stuck at this point.

I have downloaded that voltage diagram and loaded it into eagle last night, but because its my envelope that seems to be the problem, it has voltages that change over a very short period of time so I wonder how much a static snapshot of voltages is going to help me in this situation.

So we'll take this one step at a time. We're dealing here with both (fixed) DC voltages and changing voltages (like the envelopes). Let's debug the whole thing from the start and see where things get lost.

Double/triplecheck everything. Possibly use test clips with hooks and don't slip.

WITHOUT applying anything to R146 measure the voltage of the collector of Q37 (this is the middle pin). This should be around 12V. Is that correct?

Now apply the 5.33v to the "south" side of R146 (the side towards the processor). Does this change the voltage on the middle pin of Q37?

Hi MMM I'm going to cancel the break and try these steps out when I get home from work later tonight - I will update on how it goes.

P.S I must ask MMM, how did you learn your way round the xoxbox so well, i take it you have built a few or are good at following the schematics? I tend to get lost very quickly when viewing them.

[MMM]

Hi MMM I'm going to cancel the break and try these steps out when I get home from work later tonight - I will update on how it goes.

P.S I must ask MMM, how did you learn your way round the xoxbox so well, i take it you have built a few or are good at following the schematics? I tend to get lost very quickly when viewing them.

No need to rush things - I'll be posting here when time permits.

As far as my credentials go: I've been doing service work for Mode Machines since earlier this year which also involves debugging non-working Xoxboxes and fixing warranty problems. I guess I've worked on about 70 or more machines now; and I haven't found a single one that ultimatively couldn't be fixed. The design is quite clear and pretty robust.

Following the schematic: It's important to break it down into the various sections: VCO, VCF, VCA, EG. Right now we'll follow the signal through the EG (envelope generator); this is of course in the bottom left corner (the ENV part).
The Gate signal enters left through R146. Behind Q37 it is splits; one part passes through Q36 and then splits again at D34 and D35; those are Gate signals. The other part passes through Q38 and then splits through D36 (VCA Envelope) and D37 (VCF Envelope).

[maclean]

So we'll take this one step at a time. We're dealing here with both (fixed) DC voltages and changing voltages (like the envelopes). Let's debug the whole thing from the start and see where things get lost.

Double/triplecheck everything. Possibly use test clips with hooks and don't slip.

WITHOUT applying anything to R146 measure the voltage of the collector of Q37 (this is the middle pin). This should be around 12V. Is that correct?

Now apply the 5.33v to the "south" side of R146 (the side towards the processor). Does this change the voltage on the middle pin of Q37?

Okay so powered up board with J4, measured voltage of middle pin of Q37 it measured roughly 780mv 0.78v. If I apply my 5.33v clip to south side of r146 it raises the voltage of the middle pin about 3-6mv.. I wonder if that means that q37 is broken.

[MMM]

Okay so powered up board with J4, measured voltage of middle pin of Q37 it measured roughly 780mv 0.78v. If I apply my 5.33v clip to south side of r146 it raises the voltage of the middle pin about 3-6mv.. I wonder if that means that q37 is broken.

Just to be on the safe side: Where do you obtain the ground for the multimeter? Also on J4?

I think it's a good idea to remove Q37 for a test. Then measure again the solder pad for the middle pin of Q37. This should be at about 12v now.

[maclean]

Okay so powered up board with J4, measured voltage of middle pin of Q37 it measured roughly 780mv 0.78v. If I apply my 5.33v clip to south side of r146 it raises the voltage of the middle pin about 3-6mv.. I wonder if that means that q37 is broken.

Just to be on the safe side: Where do you obtain the ground for the multimeter? Also on J4?

I think it's a good idea to remove Q37 for a test. Then measure again the solder pad for the middle pin of Q37. This should be at about 12v now.

Hi MMM,

Usually I put the ground through the hole on the heatsink of the chip on the I/O board, however in this case I tried several ground points including the left side of r144 which according to the schematic is a ground... No matter which ground, I got a consistent result.

I have removed Q37 and as you said, the middle pad is measuring about 11.7v close enough I believe?... Again not going to take any further action till I hear your thoughts.

[MMM]

Usually I put the ground through the hole on the heatsink of the chip on the I/O board, however in this case I tried several ground points including the left side of r144 which according to the schematic is a ground... No matter which ground, I got a consistent result.

OK. I'm only asking because there are actually different "grounds" (an analog and a digital one) that are tied together on the IOboard. If both J3 and J4 are connected they behave as one "big" ground; if either J3 or J4 are connected (but not both) they don't which can lead to strange things.

I have removed Q37 and as you said, the middle pad is measuring about 11.7v close enough I believe?... Again not going to take any further action till I hear your thoughts.

This looks as if (at least) Q37 is faulty. If you have a spare SC536 install it. If you don't have a spare you can use Q50 from the IOboard in the meantime (you'll not have any MIDI In functionality without Q50, though).

[maclean]

OK. I'm only asking because there are actually different "grounds" (an analog and a digital one) that are tied together on the IOboard. If both J3 and J4 are connected they behave as one "big" ground; if either J3 or J4 are connected (but not both) they don't which can lead to strange things.



This looks as if (at least) Q37 is faulty. If you have a spare SC536 install it. If you don't have a spare you can use Q50 from the IOboard in the meantime (you'll not have any MIDI In functionality without Q50, though).

Do you think I should be putting the black probe somewhere else? I must admit I'm not really sure but if you have advice on the best place to use I will certainly follow. I have not even put J3 together yet and have so far only got J4.

I checked how many SC536s I have left compared to how many I need to finish the kit and I have been sent one extra in my kit!! So I soldered the extra one in and measured the middle pin and am now getting 11.7v... That's gota be a step forward eh?!