[BUILD] TB550A

[rhythminmind]

Out of curiosity.
What happens if you run the EQ into a balanced piece of gear that stays at unity while active, then into your setup.

 

[emrr]

It's an issue with the 500 rack.  I was in an Old School Audio rack, and it has some, but not all of the VPR standard jumpers.  God knows what else, haven't looked further yet.  I put the modules in my 51X rack and now I'm seeing about a 1.5 dB boost when engaged, while connected to the console patchbay. 

The AP shows slightly more than 6 dB loss in bypass. 

I did put a piece with 1246 balanced receiver and 1646 output driver in after the TB550A, and saw no change in behavior when connected to console patchbay, still +1.5 engaged.  When i put that chain in the AP I get about 1 dB more gain with eq engaged. 

 

[rhythminmind]

Ahhh I've read about the OSA racks being non-standard. Should be an easy fix.

 

[rhythminmind]

This might help.

Unlike API, BAE, and all the others, OSA racks don't have the chassis ground bussed to pin 1. I've heard a claim that they recently addressed this problem, but I've yet to come across any new OSA racks to verify it. Because of this flaw, some modules suffer poor noise rejection in OSA racks. This is because the metal enclosures of many (most? all?) 500 series modules (front panel, side panel, etc.) and/or their transformer shields are often tied to pin1 chassis ground, staying separate from pin 5/13 signal ground. This is the best way to do it, but in the absence of chassis ground, the chassis connection should probably be joined to the signal ground through one of various means.

 

[horvitz]

Well, if pin 1 didn't go anywhere it wouldn't be good but it also wouldn't account for the -4.  You would end up with buzz when doing things like touching the knobs, unless you're wearing your rubber gloves.  To my regret, I didn't account for this in my layout (won't be doing that again) so the solution would be to either hook it up in the rack or put a jumper (or small value R) from pin 1 to pin 5 on the card edge.

I've been looking around online but haven't had much luck at all finding technical info on the OSA rack.  I did find one posted picture of the backplane and it looks like there's a -2/+4 jumper for the output.  Now, if that simply selects which of pin 3/4 on the card edge goes to the XLR, it's probably not the problem because the wrong selection there would make it much worse than just -4.  Doug, can you verify that this jumper isn't doing some other weird sort of thing?

Also, I'd be curious about placing the unit with the THAT chips before the EQ rather than after.

  Brian

 

[rob61]

I have a 51x that I built. I remember when building reading about the pin 1 discussion http://www.groupdiy.com/index.php?topic=41617.msg583366#msg583366

So, instead of hard wiring the jumper as some suggested,  went ahead and installed jumpers on each. I'll check the position of those. However, I don't get any noise when touching knobs, etc, so I'm sure its jumpered for pin1 to ground.

 

[emrr]

I'm jumpered pin 1/chassis on my 51X.    No buzz when touching noted in either 51X or OSA racks. 

Putting the THAT coupled device inline so far is only affecting coupling with the AP test set, now that I'm in the 51X rack.  I'll try and have a look at it other way around, and also back in the OSA rack too. 

Have neglected to thank you for putting this project together.  They sound super useful in that broad sort of way, and full boost sounds very smooth.  No comments on the closeness to API, I haven't used one.  The bell shapes are broader than the Pultec mid EQ, and a little more narrow than old classics like the Langevin 251A. 

 

[Biasrocks]

I'm not noticing an appreciable change in level when engaging the 550 with controls set flat. I haven't measured the difference but I'm not perceiving any change. I'm using a GDIY rack along with Apogee converters in a fully balanced patchbay installation, ML2550's are the opamps w/a Classic API OT.

Regards,
Mark

 

[horvitz]

Robert, it's interesting that Doug went from his OSA rack to 51x and his -4 issue cleared up but you're having the same problem in your 51x rack (right?).  Can you tell us what else is in your path there, particularly directly before and after the 550?

I really want to be able to reproduce the problem here so I can understand what's going on and either guide folks around it or come up with some solution to be implemented on my unit.  Most particularly if I can reproduce it, I want to see if a vintage 550 does the same thing.

And Doug, you are quite welcome.  I'm really glad that people have been enjoying the EQ.  A few troubles here and there (as Mark can attest) but it's been a fun adventure for me.  Your sig is really appropriate - "Back when everything sounded great.."

 

[rickc]

could the level difference be caused by a cross-coupled line driver stage with one side floating instead of being grounded as it should?

 

[horvitz]

I'd have to think about that a little bit, but wouldn't such a device have a problem no matter what it's plugged into?

 

[rickc]

regarding cross-coupled output stages:

they depend on the receiver treating the signal as if it had come from a transformer with a floating winding.
But the proliferation of gear having the simple chain-of-inverters line driver, that imitates a transformer with a grounded centertap, requires the user to exercise some degree of intelligence when interfacing with this circuit.

Grounding the low-side of the line is permissible if your ground system can tolerate having high-current distorted waveforms going thru it. Most can't.

So the prudent way to interface with this circuit is to float t.he unused side of the output.

But the floating-balanced output (i.e. cross-coupled) circuit depends on having the low-side tied to ground, and then feedback causes the gain to change, eliminating the 6dB change caused by not having 1/2 of the output signal. You still lose 6dB of headroom, since you're losing 1/2 of the output swing. Some of these circuits demand on having the low-side of the output grounded at the source, or instability results. I'm told that there are other variants of the circuit that don't care where they're grounded, doing a better job of emulating a floating balanced transformer winding.

But AFAIK, all of these circuits can't tolerate having the low-side float when talking to an unbalanced input... The usual result is signal loss, and in some cases wonky frequency response, but not like you'd have with a floating transformer winding with one side not grounded.

For my money, the best, most universal output circuit, is a floating transformer winding. There ought to be a law against unbalanced inputs.

 

[jsteiger]

FWIW...the EQ slot in the console (where these style of EQ's were originally designed to be used) is fed from the "preamp out" which is an AP2623-1 transformer balanced output. The high side goes to the EQ input and the low side goes to ground.

The EQ module drives the (typically) 600 ohm channel fader. Naturally, this is an unbalanced component so the EQ's high goes to the fader top and the low goes to ground.

Also, besides the unbalanced/balanced nature of the source/load, the source/load impedance will also have an effect on the level the EQ passes. Since the EQ's were designed to drive the 600 ohm faders, having them drive a Hi-Z load will show an increase in gain of between 1.5dB to 2dB. None of my old EQ's are identical in this regard. Simple to check this with your AP Doug.    Also, since all the models are a little different, some drive the 600 ohm load better than others.

Cheers, Jeff

 

[emrr]

Yep, I'll say.  I will probably put in an output load resistance measured to match the bypass condition, so bypass is a useful comparison in my system.  That of course will relate to typical usage; as soon as I feed a true 600 ohm device it'll be different.  That's the only downside of switching to hard bypass.  I assume the original EQ bypass maintains the same gain through the amp, and thus gives no level shift, any level changes related to source/load being unchanged. 

It's SSLtech who doesn't believe in bypass, isn't it?  : )    I like hard bypass in things like live broadcast chains, so if something melts down it defaults to hard bypass as relays power off.  Even then, you have to hope your gain staging is correct, and you weren't forced to do gain compensation with a device that's just been reset to a different level.  I digress.....

You could get mastering-guy tweaky with audio gear and install hard relay bypass along with processing bypass that left the amp inline, and then put in the ability to calibrate amp gain.  Then you could accurately listen to source, source through amp, and source through processing for really critical decisions.  I digress further, none of that is reasonable for channel processing.......

 

[jsteiger]

Well, for kicks I tested a vintage unit and there is zero level shift when clicking the EQ in/out. Of course that is not a hard bypass. Maybe I missed but is this a true hard bypass? I haven't built one (yet!). Indeed this is the issue then. Like you have mentioned Doug, a load R switched in may handle it. I guess that is also dependent on the preceding device and the drive capability. Chances are it won't be identical to the output stage of the 550A. I suppose it can be a Pandora's box of sorts.

Brian, is there a way of modding the bypass to be like the original?

 

[horvitz]

It's funny where your decision making leads to   First, thanks Jeff as always for your insight.  I think we neglected the first rule of troubleshooting API gear - Jeff probably already knows!

Anyway, yes it is a true hard bypass.  That originally started because I just couldn't conceive of how to get the additional switching stuffed in there to do it like the original.  Then I'm sure I convinced myself "out is out" and haven't thought about it much since.  If I was starting over, I think I could do it now but we are where we are.. maybe on rev 2 since there are a few nitpick things I might like to change if I ever do it.

To answer your question, this is DIY so there's always a way, right?  The biggest problem would be finding a panel mount switch that fits.  Disconnecting the filter sections would be the easy part since you'd just snip three of the connections from the main board to the filter board and have them connected or not at the bypass switch.  Disconnecting the BP filter would involve cutting a trace and running two more wires up to the switch from components on either side of that cut trace.  The last would just be the LED which would work exactly the same as now.

So... not TOO bad actually if we could find a switch like F6UEE (precisely what's in the vintage unit) that will panel mount and fit in the available space, which is VERY tight given the HF boost/cut switch directly below.  Who's going first?  ;D

 

[horvitz]

Ohh.. I actually just had a little brainstorm.  If we cut all the traces from the relays to the ins and outs, we could use those.  It's precisely enough switching to make it work.  Hmm..

 

[rob61]

If we end up with a mod, bypassing the hardwire bypass, does that mean those of us who are getting -4db of gain reduction would always have the -4db, its just that as the EQ is switched in/out we don't notice the change in gain?

 

[horvitz]

Bring it!

Challenge!!  I suppose that's what my poor prototype is for.  The thing is already cut to ribbons, so what's a little more?

If we end up with a mod, bypassing the hardwire bypass, does that mean those of us who are getting -4db of gain reduction would always have the -4db, its just that as the EQ is switched in/out we don't notice the change in gain?

If this were to be done, it's going to be a pretty serious mod with high chances of screwing up the board real badly.  Just because of that, it will probably not be recommended.  What you'd end up with is the eq always being in no matter what but hitting the bypass would just disable the filters.  The original operates exactly this way.  If your EQ in makes -4 right now then it would always make -4.  I asked above what else is in your signal chain before and after.  You may want to try at least removing some of those things and see if it makes a difference.

 

[rob61]

In my chain, I'd tried feeding my MOTU 24i/o (with Black Lion mod) into the eq through the patchbay, then back into the 24i/o. I've also tried going interface to Serpent 4001 compressor, to the 550a and back into my interface. Both exhibited this -4db drop. Neither of those have transformers.

I do have outboard with transformers (ie Focusrite Red 2 and Red 3, Tube Tech EQs, etc) so I can experiment and see if changing what is in the chain makes a difference if that would help determine what is going on.

The EQs sound great, so no complaints here. But I like to switch in/out when comparing and the level jumps are very noticable at -4db. So a fix might be nice, but really like these EQs either way.