rhythminmind
Well-known member
- Joined
- Oct 20, 2009
- Messages
- 65
FYI - Just swapped out the JH 990's for DOA-17's. They work wonderfully in the TB550A.
[BUILD] TB550A
- Thread starter horvitz
- Start date
Help Support GroupDIY Audio Forum:
Some TB550A curves This is all the +12 and -12 settings, both peaking and shelf, and 50, 1K5, and 15K have every gain step shown.
Click this link for a full size pic
Click this link for a full size pic
horvitz
Well-known member
That's a pretty cool picture! I was never able to get them all in one place since my software only stores 10 at a time. Thanks for that.
That is beautiful artwork for geeky dorks like us! Nicely done emrr 
jimkeaney
Well-known member
Just finished my TB550 v1.1. I can't test it until tomorrow, but it sure looks like it'll function ;D Not the cleanest build (but not my worst either). I spaced-out and installed the Mill-Max sockets from the wrong side - seems like they won't be an issue as there is plenty of clearance for the op-amps, but it IS bugging me.
horvitz
Well-known member
Awesome, Jim!! Looks like you did a real nice job. Give yourself some credit there ;D
No real worries on the sockets. I have only seen one DOA tall enough to be a problem so far. Installing from the bottom is just extra breathing room.
No real worries on the sockets. I have only seen one DOA tall enough to be a problem so far. Installing from the bottom is just extra breathing room.
![Electronics Soldering Iron Kit, [Upgraded] Soldering Iron 110V 90W LCD Digital Portable Soldering Kit 180-480℃(356-896℉), Welding Tool with ON/OFF Switch, Auto-sleep, Thermostatic Design](https://m.media-amazon.com/images/I/41gRDnlyfJS._SL500_.jpg)
It will be interesting to see how they sound with the hybrid opamps.
jimkeaney
Well-known member
Thanks, Brian. Very cool kit and a lot easier than I imagined.
Yes Jeff, those are the Philips I bought from you. I have some gar2520 and API 2520's I could swap with. So I'll have to try a shootout.
Yes Jeff, those are the Philips I bought from you. I have some gar2520 and API 2520's I could swap with. So I'll have to try a shootout.
fenchelteefee
Active member
horvitz said:
Looking forward for the op-amp shootout and audio comparisons!
Cheers!
Unit7
Well-known member
Hi!
(edit: I've emailed Brian about this)
I've built Brians chorus and now two 550As and haven't had a 500 rack to test on until today. Borrowed a standard API 6 slot Lunchbox.
To my relief the chorus seems to work 100%. It's not calibrated yet. Will leave to a tech with ocsilloscope, but everything seemed to work and the chorus sounded great!
I then powered down and put the two eqs in the same slots, 1 and 2.
At power up the left channel of my monitor controller (Crane Song Avocet), fed by output 1 of the lunchbox, was going all the way up to red, and the right channel was silent. This with nothing fed into the units. I unmuted the Avocet but dispite the left ch VU going red, it was silent, at least between 20Hz-13k which is the limits of my monitoring (ADAM S2X) and my ears. When feeding audio into the units the left was still silent (still going red) but the right channel was passing audio. I did a quick check of the eq and it seemed to work fine.
I then powered down and started doing a lot of testing, powering down and up again after each attempt.
At second power up I got the problem on both units and kept having it on the 20-30 power ups after that. But problem disappears after reseating the outputs of the lunchbox, or reseating at the input of my patchbay. After that, the units seem to work properly. But at next power up the problem is back.
Problems remains regardless if unit is engaged or not at power up.
I've tested moving units to slots 3 and 4 in the lunchbox, problem remains.
I've tried unhooking my feeds to the inputs of the lunchbox before powering up, problem remains.
Until now I've been going via tielines that are connected to my patchbay. I've been using these lines a lot for years without problems, and they obviously worked with the chorus earlier.
The patchbays are Switchcraft 9625, the fairly new ones, colour blue, switches in the front for switching between no, full and half normalizing.
Also tried different cables and different inputs on my patchbays, problem remains.
I also switched between IC and unbalanced on the switches on the PCB board on the eqs, problem remains.
Finally I hooked the lunchbox up to my keyboard mixer (Yamaha 01V) and voila, it worked directly at power up!
Noted one detail at that point - When hooked up to the Yamaha mixer the units made the normal crackling noise at power up/down, which was not the case before.
If I did something wrong when building it's not unlikely I did the same fault on both units because I built them in parallel. On the other hand I never ran into the situation of having one component left which didn't match the designated place on the PCB, so it's hard to know where to start... My only hunch is it could have something to do with impedance matching on the outputs of the eq units and the receiving side in my studio.
Grateful for any ideas or advice!
Cheers
Paul
(edit: I've emailed Brian about this)
I've built Brians chorus and now two 550As and haven't had a 500 rack to test on until today. Borrowed a standard API 6 slot Lunchbox.
To my relief the chorus seems to work 100%. It's not calibrated yet. Will leave to a tech with ocsilloscope, but everything seemed to work and the chorus sounded great!
I then powered down and put the two eqs in the same slots, 1 and 2.
At power up the left channel of my monitor controller (Crane Song Avocet), fed by output 1 of the lunchbox, was going all the way up to red, and the right channel was silent. This with nothing fed into the units. I unmuted the Avocet but dispite the left ch VU going red, it was silent, at least between 20Hz-13k which is the limits of my monitoring (ADAM S2X) and my ears. When feeding audio into the units the left was still silent (still going red) but the right channel was passing audio. I did a quick check of the eq and it seemed to work fine.
I then powered down and started doing a lot of testing, powering down and up again after each attempt.
At second power up I got the problem on both units and kept having it on the 20-30 power ups after that. But problem disappears after reseating the outputs of the lunchbox, or reseating at the input of my patchbay. After that, the units seem to work properly. But at next power up the problem is back.
Problems remains regardless if unit is engaged or not at power up.
I've tested moving units to slots 3 and 4 in the lunchbox, problem remains.
I've tried unhooking my feeds to the inputs of the lunchbox before powering up, problem remains.
Until now I've been going via tielines that are connected to my patchbay. I've been using these lines a lot for years without problems, and they obviously worked with the chorus earlier.
The patchbays are Switchcraft 9625, the fairly new ones, colour blue, switches in the front for switching between no, full and half normalizing.
Also tried different cables and different inputs on my patchbays, problem remains.
I also switched between IC and unbalanced on the switches on the PCB board on the eqs, problem remains.
Finally I hooked the lunchbox up to my keyboard mixer (Yamaha 01V) and voila, it worked directly at power up!
Noted one detail at that point - When hooked up to the Yamaha mixer the units made the normal crackling noise at power up/down, which was not the case before.
If I did something wrong when building it's not unlikely I did the same fault on both units because I built them in parallel. On the other hand I never ran into the situation of having one component left which didn't match the designated place on the PCB, so it's hard to know where to start... My only hunch is it could have something to do with impedance matching on the outputs of the eq units and the receiving side in my studio.
Grateful for any ideas or advice!
Cheers
Paul
horvitz
Well-known member
Hi Paul,
So I just want to make sure that I understand exactly what your testing environment is and the results that you are seeing.
In the case where is doesn't work on power up, the output from the lunchbox goes through the patch bay and then directly into the input on the Avocet? Or, is there something else in the middle?
When you power it on in that configuration, it never works, but if you then unplug the output cable from the back of the lunchbox and then plug it back it, it always starts working OK?
Brian
So I just want to make sure that I understand exactly what your testing environment is and the results that you are seeing.
In the case where is doesn't work on power up, the output from the lunchbox goes through the patch bay and then directly into the input on the Avocet? Or, is there something else in the middle?
When you power it on in that configuration, it never works, but if you then unplug the output cable from the back of the lunchbox and then plug it back it, it always starts working OK?
Brian
Unit7
Well-known member
Thanks Brian!horvitz said:
Exactly! (on both accounts)
horvitz
Well-known member
OK, that's a bit odd and I don't have any ideas off the top of my head. Fortunately, I happen to have an Avocet right here so I will try and duplicate the situation. It will probably be tomorrow, but I'll get back to you with what I find.
Brian
Brian
Unit7
Well-known member
Thanks Brian,horvitz said:
yes, very odd. I tried both Analog in 2 & 3 on the Avocet (Analog 1 hardwired to my console). I've never had any problems with these inputs before and, as mentioned, the chorus, as everything else, worked just fine when hooked up the same way. I doubt you'll have any problems... BTW, I have GAR 2520 DOAs and EA2503 transformer.
emrr said:
horvitz said:
That was an insane number of trace cuts and re-wires! A real suicide mission, I didn't try it.
Here's another half-baked thought I'm gonna throw out to the sharks, maybe not a real solution either. It would seem the level shift condition is those cases where the balanced source is the type that wants to see a floating leg when unbalanced, rather than be grounded.
What happens if hard bypass is modified at the relays to be soft bypass, so 'low' stays connected at the input side, and only 'high' is bypassed around the unit? Output 'low' may need to swap from pin 4 to pin 5, not sure there.
Is omitting the LPF portion of the filter as simple as leaving out C3/4, or do R16/17/99 need to get tweaked as well for gain? Anyone tried that?
horvitz
Well-known member
OK, first Paul's question. You'll be happy (or not) to know that I am able to reproduce the same thing here. Very strange, indeed! At least I have something to test against, so I'll see what I can find and report back.
And Doug, you actually did that mod?? That's nuts!! ;D ;D A very brave man, indeed.
I'm not sure I follow 100% on your bypass idea. If low stays connected, it will always be grounded. Then you slip high around to either the hi or lo output, but the transformer is also still connected to both outs? Seems a bit dodgy, but perhaps I'm not really understanding. Of course, like I said before, if you are truly having an issue with this board in your studio, drop me a PM and we can talk about getting you into a v1.1. Would be a lot of work on your part, though.
As for making the filter HPF only, I would think that you remove C3/4 as mentioned, and see where you are. May or may not need to jump over R16/17. R99 definitely stays as is. Wouldn't be too hard to test out.
And Doug, you actually did that mod?? That's nuts!! ;D ;D A very brave man, indeed.
I'm not sure I follow 100% on your bypass idea. If low stays connected, it will always be grounded. Then you slip high around to either the hi or lo output, but the transformer is also still connected to both outs? Seems a bit dodgy, but perhaps I'm not really understanding. Of course, like I said before, if you are truly having an issue with this board in your studio, drop me a PM and we can talk about getting you into a v1.1. Would be a lot of work on your part, though.
As for making the filter HPF only, I would think that you remove C3/4 as mentioned, and see where you are. May or may not need to jump over R16/17. R99 definitely stays as is. Wouldn't be too hard to test out.
horvitz
Well-known member
Paul,
What we're seeing is oscillation at A2. Makes a nice huge signal at about 80-85kHz, which explains why you can see it but not hear it. I can -only- cause this to happen with GAR2520 in my v1.1 board with the output plugged directly into the Avocet. Every other combination of stuff I try doesn't exhibit this symptom. Lucky you, eh? I'm glad we found it, though.
Figuring out why this is happening might take a little longer. I would recommend that you do NOT leave the EQ on in this configuration. Probably will cook the DOA eventually.
I'll be back with you when I have something.
Brian
What we're seeing is oscillation at A2. Makes a nice huge signal at about 80-85kHz, which explains why you can see it but not hear it. I can -only- cause this to happen with GAR2520 in my v1.1 board with the output plugged directly into the Avocet. Every other combination of stuff I try doesn't exhibit this symptom. Lucky you, eh? I'm glad we found it, though.
Figuring out why this is happening might take a little longer. I would recommend that you do NOT leave the EQ on in this configuration. Probably will cook the DOA eventually.
I'll be back with you when I have something.
Brian
Brian, for giggles, try a load R after the 2503. Something around 1 to 2k-ish from hi to lo. I have seen a major ringing like this on an old 553 I had here a while back. Only one out of the 3 had this ringing problem. Sufficient loading cleared it right up. Don't know what the input Z is of this Avocet thingy you guys are talking about but keep in mind the original API EQ's were designed to drive a 600 ohm load in the consoles.
Sorry, I see my vague wording. No, I didn't try your bypass mod. It was insane. Really for the better that it's not posted any more!
Low being connected to ground at input, then disconnected by bypass seems to be the root of the problem. If low stays connected at input, then the gain condition shouldn't change. I think output low would have to move from pin 4 to pin 5, or the relay short them together in bypass state. Then bypass should reflect whatever the input conditions actually are, whereas currently a V1 build will cause some sources to change gain switching from balanced to unbalanced. We still won't be running audio through the amp.
Hang on, even easier. The input doesn't have to switch at all. Soft bypass need be nothing more than parallel path from the input (pins 8/10) going to relay, relay switching either output transformer or input parallel path to output pins. Input condition stays the same, output is either the amp, or the parallel input path.
Low being connected to ground at input, then disconnected by bypass seems to be the root of the problem. If low stays connected at input, then the gain condition shouldn't change. I think output low would have to move from pin 4 to pin 5, or the relay short them together in bypass state. Then bypass should reflect whatever the input conditions actually are, whereas currently a V1 build will cause some sources to change gain switching from balanced to unbalanced. We still won't be running audio through the amp.
Hang on, even easier. The input doesn't have to switch at all. Soft bypass need be nothing more than parallel path from the input (pins 8/10) going to relay, relay switching either output transformer or input parallel path to output pins. Input condition stays the same, output is either the amp, or the parallel input path.
horvitz
Well-known member
jsteiger said:
I forgot rule #1 of troubleshooting API gear: Ask Jeff because he's probably already seen it. I dropped in a 1.5k across the output and it went right away. But, I'm still scratching my head a little as to why my v1 board does not exhibit the same symptom. I'm going to have to tear into it and see what's up.
Thanks!
