[BUILD] TB550A

[horvitz]

Erez,

Glad everything got to you ok!

First on the schematic, you can find that in the tech docs section in the API category.  That schem is correct for my implementation with the exception of one resistor in the C-deck of S8 (I use R46 rather than R47, which I think was a typo on the original schematic).

As for your troubles, I always start with the amps.  I assume that you've swapped op amps from the working unit to the non-working one?  To start to narrow down the problem, you'll want to inject a signal and check the output of each DOA and all four DVFs to see where it's falling off.  If you find one that isn't good, also check with your DMM for any significant DC at that point.  Let me know what you find with this and then we'll drill down into whatever section of the circuit is not working correctly.

  Brian

 

[whomper]

Brian,

Thanks for the hints.

As soon as I activate the circuit with the on switch, the input signal disappears from input and I find it hard to test. It is like there is a short circuit when on that nullifies the input signal.

Perhaps the transformer messes up the input? For this I had even removed the op amps but it behaves the same. Perhaps I should remove the DVFs as well and test?

 

[horvitz]

Now that's interesting, indeed.  I wouldn't say right now that it's the transformer or DVFs.

Let's start at the very beginning.  With the op amps out (at least A1), and the circuit in, check for a signal at pin 10 on the card edge, each side of R1, and the in+ of A1.  You can see that path pretty clearly in the top left corner of the schematic.  From what you're describing, it seems like it's not even getting that far.

 

[whomper]

That is the strange thing. I am checking at the XLR in connector (Have Igor's Rackit with the two XLRs) and the signal evaporates as soon as I activate the circuit, like it is nullified with ground somehow.

To be noted, I do not have the filter board and non of the op amps installed.

Another thing, A3's trannies run very hot. I am about to remove it from the board and test again.

Edit: Removed A3, but no change.

 

[horvitz]

OK, let's check a couple of things.  With the unit off and disconnected, check continuity from the hi in (pin 10 on the card edge) to the side of R1 closest to the rear.  Also check continuity from the other side of R1 to the + In on A1.  Additionally, check resistance from pin 10 on the card edge to pin 5.  You should see about 18.5k there.

Regarding A3, it certainly shouldn't get hot, but I don't think we're even getting that far so let's not worry about that just now.  And yes, leave the filter board and op amps out for now.

 

[whomper]

All tested as should (both continuity and resistance). Have also disconnected the red/brn from the board, but no difference.

Have noticed that the strange behavior is happening whenever I connect the power supply analog ground to the board IN gnd connector. Not sure why it is related.

Hmmm, next I'll probably take apart the working unit and will try to compare then electronically. Perhaps there is some short circuit somewhere.

 

[horvitz]

Are you saying that the problem doesn't happen with something disconnected on the power supply?  Things that should eventually get to the PSU ground are pins 1, 5, and 13 on the card edge.  Pin 1 is the chassis gnd and although it should be connected, the unit would work without.

 

[whomper]

Ok, can report some progress.

The case of the missing signal is no longer an issue. Due to cable and PSU problems, I had grounding issues. But that is resolved.
I can engage the unit and trace the signal. With only A1 installed (not A2), I am getting signal at the A1 OUT.

A3 2n5087 is the one getting as hot as hell the second I power the unit, which makes it difficult to perform long traces...

A2 IN does not get any signal.

What should I look for now? I find it difficult to understand the schematics as I am not sure from where should A2 get its signal.

Thanks for all the help so far

 

[horvitz]

Progress is good!  It's always a pain when we're looking at multiple issues.  Sounds like we're getting there, though.  So if A3 were not passing signal, you wouldn't see anything on the In+ on A2.  If you go backward on the schematic from the A2 In+, you'll see it go through the cut side of the A deck on the LF boost/cut switch and then to the output from A3.  Yes, the schematic is terribly confusing!

Let's then have a closer look at A3.  Of course, inspect it closely for any bridged or bad solder joints.  The 5087 which is getting hot could certainly be bad as well.  Maybe that's something you could source locally?  I'll be happy to send you anything you need, but the cost of shipping would be enough to buy 100 of them and will take a while.  I think the easiest thing to do might be to grab a DVF out of your working unit and install it here to see if the problem stays with the unit or follows the DVF.  In that way, we can know for sure where to focus our effort.

Let me know what that does for you and we'll take it from there.

 

[whomper]

Switched A6 with A3 and sure thing, the problem moved to A6, hence it is probably the DVF. Changed the 2n5087 with a new one. Still same behavior, so it might be one of the 2n5088 as I checked the circuit and it seems kosher.

As I can not seem to source the 2n5088 here, can I use a replacement? What about a BC549 or something similar. Will need to take care of pinout, but perhaps it will behave close enough, or must I use the 5088?

The working unit is already reassembled, so am looking at ways not to disturb its working condition...

 

[horvitz]

Well, I haven't ever tried to build one with BC549, or anything else for that matter.  You can't make it worse, right?  Might be worth a shot anyway.  If you'd rather, I'll be happy to send you the parts and a PCB to build a whole new DVF although I imagine that it will take a few weeks to get there.  Let me know if you'd like me to do that.

 

[whomper]

From what I have been able to read, the 2n5088 have a hfe of 800 while the BC549 have an hfe of about 400. Does it make a difference? Have read somewhere that it should provide a lower noise alternative to the BC549, so I would probably look to have the original parts sourced.

Will probably go off grid till I have the parts and test.

 

[whomper]

Brian,

Many thanks for your help. The replacement parcel came in fast and I was able to build another DVF board in no time, plugging it in and get good sound out!
Greatly appreciate the high service level I received! It is was makes a difference.

Keep up the good work.

 

[qmp audio]

finally the tb550a came up in the cue!  just finished the build... now to test.

 

[horvitz]

Very nice, sir.  I like the blue knob inserts!

 

[qmp audio]

Very nice, sir.  I like the blue knob inserts!

Thanks!  Here it is all racked up:

 

[horvitz]

So.. been thinking (I know, bad start already).  I'll do another run of these hopefully around the new year, depends how fast the TBDD sells.  Maybe with that, it would be nice to do a rev B of the board.  Aside from some minor layout changes, I have a few things in mind that might be good and thought I'd ask for thoughts and opinions from the rest of you on what you might like to see.

In no particular order, here's what I'm thinking:

THAT Input. - Op amps would stay as normal, but this would just sit in front.  It would be optional of course.

THAT Output - Anybody care to have a less expensive option for output??  I wouldn't like it without the transformer myself, but I don't think this would be hard to add if there's interest.

Soft bypass - ditch the relays altogether and do bypass just like the original.  It would actually be more like the 550 than
550A but they are same enough so it wouldn't matter to us (I think - needs testing).  This would probably be required anyway to make room for the IC in/out options.

Different caps?  The Nichicon YX may be a nice choice for this EQ.  Would take out about 10-15 caps too but not sure if they'll fit due to the larger format.  Would still use the Xicon polystyrenes for the small pF values.  I'm pretty happy with the WIMA option myself and am inclined to leave it that way.

Any other thoughts?  A whole bunch of you have built these so I welcome any feedback regarding the construction process or whatever you've found in using the unit.

Thanks! The support from everyone on this project has been awesome so far.

  Brian

 

[emrr]

#1: original type bypass.  The hard bypass becomes a useless option when gain doesn't match exactly for both states, and I've found it really depends what I'm patched into as to what the gain difference is.  Even with #2 option below added, the output load can make gain move enough to make bypass a non-trustworthy comparison tool.  It doesn't take much of a dB for level differences to change perception of EQ changes. 

#2: THAT input for those concerned about level shift in varying situations.  I've considered adding a daughter board.

I wouldn't consider THAT output (A2/T1 comprise a system) or worry about different caps. 

 

[qmp audio]

I would definitely be interested in a pair or 2.  I'm enjoying mine but i only have one unfortunately.

I do like the hard bypass.  I have mine setup half-normalled to a patchbay channel strip ( pre - eq - comp ) and like the option to fully remove the EQ if needed w/o patching.
I have noticed a DB drop when using as an insert.  No DB drop when using in a channel strip config.

Can't wait for round 2 whatever you end up doing.

 

[horvitz]

Thanks, guys.  Good feedback!

Space allowing, I may be able to make the hard/soft bypass user selectable.  Or perhaps something like on/off/out from the front.  I'll think about it.

I'd like to potentially offer a minimal kit with this so that existing users could salvage the expensive bits from their current unit and build a newer one if they really have the desire.  This would of course require no front panel changes, which I'd certainly like.