[BUILD] TBDD

[horvitz]

Great!  Let us know how it works out.

Thanks, Udo

 

[fazeka]

Proud papa showing off his newborn twin TBDD rack units. I certainly hope more people build the rack version. Nice compact 1RU form factor.

Thanks to Brian for making it happen! ;D

 

[okgb]

Looks Great ! It's my plan to do a rack one, but I'm trying to be responsible
and finish a few other things first !

 

[horvitz]

Awww yeah!!  I'm so happy to see those done and working.  Really, really nice job.

And Greg, I'm not sure I like this whole "responsibility" thing.  We'll not have any of that talk around here.

 

[jplebre]

wow you really did give it some "space" seems so much bigger than the 500 version

 

[fazeka]

Sadly, I've jumped off the 500 series bandwagon. 8) Nice to have options for non-500 series fans.

 

[okgb]

But more compact than the original 2RU , but really , How are you going to get another dimension without space ?

 

[rob_gould]

Great!  Let us know how it works out.

Thanks, Udo

4

Not so well, annoyingly.

First impressions were good.  I swapped the 1646 and 1246 in the right hand (clicky) side, turned everything on and the clicking had gone.

So I started calibration again.  Once again both bias' were set at -7.5V - both TP12 and TP22.

I broke out my old HP scope and checked that the sine wave going in looked like the sine wave coming out, which it did on both channels but the signal was a little lower (1.55V on the way in, about 1.47V on the way out on both sides)

Here's the clean sine wave on pins 2 & 4 while in soft bypass mode - looks OK.

And here's a short vid of what I see at TP13

I thought that this should be the same sine wave as I see at the output shouldn't it?

And here's a short video of what I get at TP13 with the chorus engaged and no input, trying to examine the clock signal

Surely this doesn't look right either does it?

To measure these I'm using the test points, and grounding at pin 5 of my 18 pin extension lead from the 51x rack.

Also worryingly I can't see anything on TP23.  It doesn't even trigger the scope when I probe on TP23.

I'm beginning to wonder what I could have damaged on both sides of the unit  down circuit of the 1646 and 1246 when those chips were in the wrong positions.

I've also pulled the whole thing to bits and checked all chips for orientation and to ensure they're in the sockets properly, that transistors are in the right places and orientated correctly, I've checked all the electrolytics and as many of the polys as I can read the values of without desoldering them, and had a quick check over resistor values.  I'm pretty happy that the component placement is OK.

Do you have a recommendation of ways to trace the signal round the circuit to try to find out where it's going wrong?  Otherwise I thing it might be worthwhile just doing a complete swap out of all the chips.

Another question : I note when I turn the wet / dry knob that the output signal is much higher in the middle (50 / 50) than when fully wet or fully dry.  Is that normal?

 

[horvitz]

OK, so not horrible.  Good news is that we put any THAT issues behind us and can now have a good look at the rest of the circuit.  I think you're not far off here!  I'll try to address your questions and then we can figure out what's next.

I thought that this should be the same sine wave as I see at the output shouldn't it?

Actually, no.  TP13 is the output of the first op amp following the delay.  The dry signal that makes up the soft bypass (and the dry side of the blend) comes from before this point.  So while it isn't being used at the output, the signal still runs through and gets processed.  That kind of fuzzy signal that hops around is exactly what we expect at that point.  So, looks good to me.

And here's a short video of what I get at TP13 with the chorus engaged and no input, trying to examine the clock signal
Surely this doesn't look right either does it

It doesn't look bad.  The clock signal seems right.  It stands still when looking at it with no signal and chorus off?  When you engage the LFO, your scope may be having a hard time locking on.  On mine, I can see it kind of accordion in and out in time to the LFO.  If need be, I can try to take a video of this later.

To measure these I'm using the test points, and grounding at pin 5 of my 18 pin extension lead from the 51x rack.

Should be fine.  I just clip ground to chassis but that should work too.

Also worryingly I can't see anything on TP23.  It doesn't even trigger the scope when I probe on TP23.

I'm beginning to wonder what I could have damaged on both sides of the unit  down circuit of the 1646 and 1246 when those chips were in the wrong positions.

This is the only thing that you've mentioned so far that seems wrong.  Can you please swap the MN3101 chips from left to right and see if the problem follows?  A dead clock chip would do exactly this.  Happy to send you another if this is the problem.

I don't think you've damaged anything at all further than those chips that you replaced.  Even though I test about 20% of the BBD and clock chips before they go out, there have been other failures.  When we get to the next round of these, I'll be testing 100% to try and head this off better.  They are old stock, so it is bound to happen.

Another question : I note when I turn the wet / dry knob that the output signal is much higher in the middle (50 / 50) than when fully wet or fully dry.  Is that normal?

Much higher, or a bit higher?  This was a concession to cost and available space.  If you set to soft bypass so you are blending dry/dry, the blend will make about +3dB at center.

So to sum up, the only thing I've heard here that doesn't sound good is the lack of clock at TP23.  Try swapping those chips and let me know what you get.

  Brian

 

[rob_gould]

Hi Brian,

Well, with some good advice and not trying to troubleshoot so late at night I'm getting somewhere

This is the only thing that you've mentioned so far that seems wrong.  Can you please swap the MN3101 chips from left to right and see if the problem follows?  A dead clock chip would do exactly this.  Happy to send you another if this is the problem.

I don't think you've damaged anything at all further than those chips that you replaced.  Even though I test about 20% of the BBD and clock chips before they go out, there have been other failures.  When we get to the next round of these, I'll be testing 100% to try and head this off better.  They are old stock, so it is bound to happen.

The problem does indeed move when I swap this chip.  I can now measure the clock on TP23...

It doesn't look bad.  The clock signal seems right.  It stands still when looking at it with no signal and chorus off?  When you engage the LFO, your scope may be having a hard time locking on.  On mine, I can see it kind of accordion in and out in time to the LFO.  If need be, I can try to take a video of this later.

Something like this?!  (bypass, chorus 1, chorus 4 - you can hear the click of the switches over the noisy fan of my old school 'scope)

Much higher, or a bit higher?  This was a concession to cost and available space.  If you set to soft bypass so you are blending dry/dry, the blend will make about +3dB at center.

Performing the test you suggested, it's bang on 3dB higher in the middle


So in summary, your diagnoses look completely correct.  Thanks for your assistance in troubleshooting this; I really appreciate it. 

Please may I take up your offer of a replacement MN3101?

 

[horvitz]

Something like this?!  (bypass, chorus 1, chorus 4 - you can hear the click of the switches over the noisy fan of my old school 'scope)

That's exactly right.  Doing precisely what we expect.

Please may I take up your offer of a replacement MN3101?

No problem at all.  I'll get one tested and out today.

 

[jplebre]

That info is gold Brian! I feel like I learnt a bit more about the circuit with your exchange!
Cheers

 

[rob_gould]

Please may I take up your offer of a replacement MN3101?

No problem at all.  I'll get one tested and out today.
[/quote]


Thanks once again!

Rob

 

[Rob Flinn]

I just finished constructing the rack version of this unit.  It passes & signal, & there is a difference when I switch it in & out.  However I'm not entirely convinced the LFO part is working. 

The reason I say this is because I used to have a Roland Juno 106, which had a chorus that I think was similar(ish) to the DIM D, & switching it in & out there was quite a dramatic difference.  i.e the synth sounded a bit weedy without it.  I'm not hearing that sort of difference.  I think perhaps the signal is being passed through the BBD's but not much chorusing is happening.

I don't see the LFO modulating the clock with my scope, & I don't see a double speed on settings 3 & 4 compared to 1 & 2.  Which point on the LFO board should I be able to see the LFO wave form ?

 

[horvitz]

Rob, have a look at pin 1 on the 4558 on the LFO board.  You should see it hop back and forth between about + and - 14V.  In modes 1 and 2 it will flip about every 2 seconds and in modes 3 and 4 about every 1 second.

  Brian

 

[Rob Flinn]

i can sense it pulsing, but it doesn't seem to go positive.  It seems to sit at about -14v & pulse momentarily off then back on at about the tempo you say.

On the clock test point I get the waveform below, but it is absolutely static, even if the LFO is switched in.

 

[horvitz]

Your clock doesn't look terrible.  I've attached a picture of a calibrated clock output.  It's a very small signal that you're looking at.  I'm set to 100mV and 5us/div in this picture.  It won't always look quite this good.  Sometimes you'll see every other one a little bigger or smaller, which is just fine.

You're seeing the same thing on both left and right channels?

The next place to look at the LFO is at pin 1 on the TL082.  In modes 1 and 2, you'll see it moving smoothly between about +/- 3.5V, modes 3 and 4 between about +/- 2V.

A note if you're looking at the schematic (in the tech docs here), pin 7 on the 4558 on the schem is pin 1 on my version.  The two halves of the TL082 are swapped, as well.  Both of the test points that I had you look at are indicated on the schematic.

There are a couple of other interesting test points for the clock signal if you look on the schematic up around the MN3101.  All of my part numbers in the main section of the board and the LFO are the same as theirs, so it should be easy to see what's what if you're poking around.

 

[Rob Flinn]

No, the LFO isn't playing ball at all.  I have to admit to having a TL092 in place of the TL082, because I couldn't get one until tomorrow.  Although I would have thought that this wouldn't have made a massive difference.  I'll grab the 082 tomorrow & check it again.

 

[horvitz]

I would have guessed the same with the TL092 but I haven't tried that myself.  Unfortunately, I don't have one here to check it out.  Let me know how it goes.

 

[Rob Flinn]

All working now.  I think the issue was that the 092 had quite tarnished pins which I probably didn't notice without my glasses or something like that.  Anyway popped an 082 in & it's all working apart from the remote bypass jack now.  But I don't think that is a feature I'm going to be using so not too much of an issue.  I'll post some pics up if you're interested.