[BUILD] TB550A

[Unit7]

Great!  Glad you got it worked out.  Unfortunately, that's kind of an easy mistake to make but that's why we have the sorting sheets.  Real fast too.  I think you give Unit7 a run for his money on build speed!  8)

;D Well, I'm not fast...just manic...did my last one in three movies incl commercials between 7pm to 2am, and then the filter board and final stuff in four hrs the next day, so approx 11hrs total, excl the DOAs. And despite that I use the sorting sheets I've also mixed up .001µF and .0015µF a couple of times. After a few hours  of work and the sorting sheets flooded with components the '5'  sometimes gets mixed up with the following letters. The '15' on my sheet is now underlined... Luckily always noticed before moving on to the next cap though.

 

[cbarrett92]

I had them sorted, if only my hands would grab what my brain was telling them to grab!

The build took me maybe 9-10 hours, which is well worth it for such a quality piece of gear that you end up with in the end.

 

[Dr Gris]

Hi folks,

Don't know if I'm entitled to any help here because I didn't build these myself,
when I finally had the dough to buy kits the last ones of the first batch went quick.
So I bought two of these lovely EQ's from the black market.

My problem is that in one of the modules the high shelf doesn't work, only peak.
I've reheated the solderjoints around the switch and measured with a DMM and the switch seems fine.
Don't know where to to from there...
And I can't find a schematic for these.

Love your EQ' s and they look really great!

Best regards
Magnus

 

[horvitz]

Hi Magnus,

We can certainly get you sorted out here.  The good news is that there is very little of the circuit that is different between peak and shelf, so we should be able to narrow it down quickly.

First, I'd recommend grabbing the schematic.  You can find it in the tech docs section under API.  There are a few versions out there, but that one is correct for my implementation.  Also if you PM me your email address, I'll get the v1.0 build docs out to you so you have some reference.  It will be much easier to talk about testing if you have the board layouts in front of you.

The very first thing I would look at is to make sure that the signal is getting to the right parts of the filter board when you flip the peak/shelf switch.  Inject a test tone and then of course you'll need a scope or signal probe to go looking for it.  We have three places to check.  You'll notice that the HF switch is a 2-pole, 2-deck switch but we are actually only using three of those.

First have a look at the A common terminal, which is on the rear deck toward the inside (the bottom as the unit would sit in a rack).  That terminal should always have signal regardless of the Peak/Shelf switch position.

Next, we're looking at the B and C common terminals.  B is on the other side of the rear deck and C is on the front deck right next to A (don't worry, there is a picture of this in the build guide).  When the switch is in peak mode, you should see the signal on the common terminal of B.  When in shelf, it should be on the common terminal of C.

Let me know what you find with that and we'll take it from there.

  Brian

 

[horvitz]

Oh, and I just want to verify that we are talking about the v1.0 PCBs - the black boards?  I kind of inferred that but don't want to assume..

 

[Dr Gris]

Hi Brian,

Thanks for helping me!!
I guess I've been lucky so far cuz all my own builds has worked right away  8)

The boards are from the v1.0 PCBs.
Just got a used old Tek 2225 scope and a probe I'm trying to get a hang of.
As I'm a tech dumbass newb it would be nice if you could mail me the build guide.
My email is shown below my user name.

As I'm very busy at the moment I apologise if I'm not being able to respond right away.

Respectfully
M

 

[horvitz]

My email is shown below my user name.

Maybe I'm missing something?  I don't see it.

 

[Dr Gris]

You can click on the envelope sign below the user name if the member has allowed his email to be shown
to other members when they are logged in. I can see you have that activated. Others don't.

Anyway, my mail is: magnus dot norrenberg at comhem dot se

Best
M

 

[Dr Gris]

Ok, I did some tracing with my scope and there's signal on common terminal B in peak,
and terminal C in shelf.

Just edited this reply 'cause I mixed up terminal B and 1, oops...

Best
M

 

[horvitz]

OK, email coming.  I think if I hit that envelope icon that I can't do an attachment, so I couldn't sent the guide that way.  Anyway, coming in an email in just a moment.

So if you are seeing the signal on the C common in shelf mode, the next thing to check is to make sure that it's getting back to the boost/cut switch.  For that, you'll need to remove the unit from the front panel and bracket.  On the boost/cut, there are four decks.  They are A-D from the front to the back.  With it in shelf mode, you should expect to see a signal on the common terminal on the C deck, and then it should not there in peak mode.  Let me know how you do with that.

 

[Dr Gris]

Thanks,

I got the email. Tried some probing but only got weird traces on the scope exept for when I probed
the trafo, then I got a clean sine wave.

I'm tired and just got home from a gig, I'll try to get some time for it tomorrow morning
before I leave for some other gigs this weekend.

A big thanks so far!

//M

 

[Dr Gris]

Hi again,

Probed some more today. If I'm doing it right it is as follows:
Common C deck on boost/cut switch: no signal in shelf (or peak)

Maybe totally unimportant but for comparison:

Common: A deck: full signal in shelf/peak
                B deck: less signal in peak and about half of that in shelf
                D deck: full signal in peak and a tiny bit less in shelf

Measurements taken with Boost @ +2dB, frequency @ 5kHz
Testtone: 1kHz @ 0.775V rms

Best
Magnus

 

[horvitz]

OK, I think we're getting close.  One question that I didn't ask is, do you get the same result across all frequencies on the HF?

If you'll look at the schematic, notice switch S9C on the lower left.  That's the frequency select for the HF shelf.  We already established that in shelf mode, signal is getting to the common terminal on that switch.  From there, it goes through the appropriate caps for each freq. and then on to the common terminal on switch S8C, where you just told me it isn't getting to.

If the problem exists on all frequencies, it's probably not a cap.  The next thing I would look at is the path from the far side of the caps down to the common terminal on S8C.  So, inject a signal in shelf mode again and start at the 5-pin connector at the rear of the filter board just behind the HF switch.  You should see the signal on pin 3 (the center pin).  If you do, check the adjoining pin on the main PCB and make sure it's getting from top to bottom.  There is then a trace from that point straight to the C deck common terminal on the HF boost/cut.  I would think that it is broken in there somewhere.

  Brian

 

[Dr Gris]

Ok, same result on all frequencies.
There's no signal on pin 3 (center pin) on the header. That goes for both the filter board and main PCB joints.
I have continuity though.

Thanks again!
Best
//M

 

[Unit7]

Discovered a problem today with the DIY 500 rack in combination w one of my TB550As.
Edit: Problem was DIY PSU, not TB550A.

 

[horvitz]

Ok, same result on all frequencies.
There's no signal on pin 3 (center pin) on the header. That goes for both the filter board and main PCB joints.
I have continuity though.

Interesting.  I sort of wonder if the Grayhill switch is broken.  Very uncommon, but I've certainly seen it before.  Let's do a continuity test to check that out.  With the switch set to 5k, you should have a connection from the C deck common terminal to the 7th pin from the top on the front deck.  7k->pin 8, 10k->pin 9, etc.  Let me know what you see there.

  Brian

 

[horvitz]

Discovered a problem today with the DIY 500 rack in combination w one of my TB550As. The signal sounded very weak and thin thru this module. I swapped modules and problem moved, so the cables/500-rack seems ok. But when I put the same 550-module in my Purple 500 rack it sounds good. Any clue? Only thing I come to think of is that the DIY 500 rack doesn't allow the modules to sit as deep into the connectors as in the Purple rack. Posting in both the DIY rack and TB550A build threads. Grateful for any advice!

So you're saying that 4 of 5 work correctly in the GDIY rack and just one doesn't?  If that's the case, I certainly wouldn't think that it's any physical issue with the rack.  There are lots of these out there in GDIY racks and no problems.  If it wasn't seating far enough in, it wouldn't work at all as opposed to weak and thin, as you describe it.

What I'd pay attention to are the differences between the racks.  What gear is before and after?  IC vs. unbalanced input - is it set the same as the others?  Try both modes and see if it improves with one or the other.  Perhaps swap DOAs with one of the working ones?  If there is a difference in what the output is driving and there is a problem with A2, that could do it.  Did you swap it over to the Purple rack and move the cables with it so the chain is exactly identical with nothing at all changed except for the rack?

  Brian

 

[Dr Gris]

Well, the good news is that the Grayhill frequency switch is ok.

I must admit, my brain seems to be on holiday too. This shouldn't be so hard...
When I look at the schematic I think I get the signal chain, but then when I turn over to the module I feel lost...

Thanks!
//M

 

[horvitz]

You know, that schematic is actually more confusing than it looks!  It's a very tricky circuit.  We're actually down to the part where I diverge from the original a bit with all those caps on the filter board.  This was the concession (not a bad one, I think) to not having the custom switches that API used.

So, since we've really narrowed it down to a very small area.  What I need to do for you is come up with a testing path right through the filter caps on that 5k setting.  Unfortunately, I think it will be tomorrow before I get to that.  I'll put together a picture of point A->B->C->etc., so we can figure out exactly where it's falling off.

  Brian

 

[horvitz]

OK, first please forgive the outstanding crappiness of this picture.  My camera broke so that's with my cell phone, and also the only v1.0 that I have is a prototype so it's seen better days!

That caveat out of the way, this picture shows where you want to find your signal on the HF in shelf mode at 5k.  All the red dots are where the signal should be.  The lines represent the things that should have direct continuity.