[BUILD] 1176LN Rev D DIY

[jplebre]

Forgot to say first time around, the QBias trimmer was a goner. Had to swap it. Not sure how often this happens but I think that's why Mike was asking you for the voltage?
correct me if I'm wrong Mike

 

[Idiophonic]

Idiophonic, Q7 Q8 Q9 and Q10 are NPN BJT. The legs are collector, base and emitter (top to bottom, when reading the schematic).
If you can confirm you simply swapped the names (which seems to be the case) then your voltages are on the ballpark http://mnats.net/files/1176REVD_VOLTS.pdf

Funny, I just built a second PCB and the threshold is pretty much where you described, around 7-8 o'clock.

Hi jplebre, thanks for the reply.

Well, it didn't take me long to out myself as a newb! Of course those values should have been labeled base, collector and emitter.  I measured the FET at Q1 first and then didn't change my nomenclature... D'oh!

Anyway, I had a chance to compare my unit to a UREI and UA version with the help of a highly respected tech here in NYC. 

My unit had more preamp gain, more output amp gain, and a lower threshold than the other two units.  Measuring voltages at different points in the circuit, we did not find an obvious or simple explanation for any of these issues. 

We found some differences in the circuits.  There are essentially points in the UREI schematic that have extra resistors in parallel labelled T&C (testing and calibration?).  It was explained to me that these resistor values would be chosen during testing and calibration to achieve the desired amount of preamp gain, for instance.  We decided to install variable trimmers in certain locations to allow us to easily match the behavior of my unit to the others.

The mods were a success; my unit now has similar input gain and threshold to the others.  We did not modify the output amp yet.  Everything sounds fantastic, i just have the output knob below 12:00 most of the time. 

As for the output amp, is it possible the tertiary feedback winding on the output transformer is not providing the negative feedback it should?

Anyway, I am happy, my unit was closer to the UREI than the UA was!

 

[sage]

it would be interesting if you could let us know where did you install the variables - ultimately it is about how it sounds but certainly getting it as close as possible to the original could be the goal...

sage

 

[jplebre]

Interesting!
Yes the TC points where points "added in factory" according to the manual and other sources.

If you have the chance, can you please add some notes on the schematic? even if it's some MS Paint or some Gimp skills is fine
Pretty intrigued.

Put together 2 PCB and 2 Attack/release constant sections and having a similar problem, using components from same batch and similar values. There's a T+C value next to the QBias that it is "empty" on these versions, wandering if you did work on this section?

I need to make my attack faster by 30µsec and slower by a more than a few 100µsec

 

[sage]

IMPORTANT TO ALL CINEMAG USERS:

I finally put my units (REV A and REV D) in the rack in the studio and we checked the PHASE and realized the units were both outputting out of phase...

Obviuosly we checked the XLRs and the Patchbay and it's all good... so we had to swap the polarity on the INPUT trafos - so now the Yellow and Orange are reversed AND THIS FIXED THE ISSUE !

So if you used the CINEMAG TRAFO you MUST WIRE THE WAY SHOWN IN THE PICTURE:

NOW IT SOUNDS AND GREAT AND IN PHASE 

sage

 

[Che_Guitarra]

I'm about to jump on the Hairball train and build a 1176 myself.  Just doing a bit more research to satisfy myself that I want a Rev D over a Rev A.

I've found the Hairball/Mnats parts list and Mouser link and just want to ask - if you look at the front page of Mnats PDF here: http://mnats.net/files/DIY1176LNREVD_V2.pdf, i'm seeing a lot more carbon comp resistors than I am on the PCB of other builds (such as DigitalMetal's post #2602 of this thread - page 131).  I understand the noise floor may raise if I opt for more carbon comp resistors in the circuit - but is there a 'vibe' payoff using these resistors?  I ask this as I am wanting muchly to emulate the Waves CLA 'Bluey' plugin, which sounds ever so slightly looser/warmer than the average Rev D. 

Sorry if this has been asked/answered here before, but i'm a newbie to the forum and flicking through threads with hundreds of pages is a bit daunting 

 

[RichardM]

Hi Guys,

Just completed my 2nd rev d. My first attempt didn't go so well but all of my previous issues with the first build were solved with experience. Everything works, it is very quiet and it sounds great but I am having trouble with the calibration. The problem I am having is with the qbias adjustment. I am using the calibration video at the end of the rev d wiring pages and I can't get the qbias to make the 1db drop during the calibration process. I can't seem to get it to do anything. Is there something that I could be missing?Thanks

Rich

 

[jplebre]

Rich, you might be too far out of the operating region.

If you had to really crank it up to get +1, you are probably all the way to ground (left on schematic). If you were in the ballpark with the video or lower, you are probably far too high to the right. The VIshay blue trimpots have this stupid amount of turns you might have to give it a couple of twists before you bring it to where it should be.

DO you have your meter on +4, attack off (clicked off or GR jumped), release fast (all the way clockwise if you are being faithful to the design) and ratio on 20:1?

If all the above fails, check your voltages agains the mnats schemo and we can try and find the culprit

Cheers

 

[sage]

From a previous POST of Mike at Hairball, this helped me so much:


For setting the qbias try this:

If you have a version with the "off" switch on the attack pot, you can just switch that.  That will ground pad 22.  Feed a 1khz signal into the input at about 0dBu.  You can put your DMM across input xlr pin 2 and 3 and look for 0.775VAC which is about 0dbu.

Now put your DMM across pin 2 and 3 on the output xlr and turn the qbias (in and out knob mid way).  You should see the output moving up or down.  Turn the bias until it gets to the max and doesn't get any higher. 

Then adjust the output knob so you see +11dbu (2.75VAC) at the output.  Now pull the qbias back until you get a 1dbu drop 10dbu (2.45VAC).  That will set you bias.

 

[RichardM]

Hi Again,

I just can't seem to get any change from moving the q bias trim pot. I have tested some of the voltages and the 30v at the side of r79 tests exactly as it should but many of the others I think are too high. G at q11 = -1.8 vdc    D at Q11=18.3 vdc and S at Q11= -2.3 vdc.  I am really unsure what to do because I thought that the voltages in this region would not read as they should until the unit is calibrated properly. The o adjust trimmer that is accessed in the front works as it should and I am sure that I am following the calibration steps properly. Do you have any suggestions? My only other concern at the moment is that the output knob raises the level as it should when being turned up but as it nears the max, it makes a bit of a clunk and enters a hiss zone. Maybe this has to do with the calibration, but I am unsure. Thanks for all your help!

Rich

 

[jplebre]

My only other concern at the moment is that the output knob raises the level as it should when being turned up but as it nears the max, it makes a bit of a clunk and enters a hiss zone.

Did you use shielded cable for the connections tot his pot? If so and still persists, this issue was solved for me when I grounded the board. Solder a wire to the negative side of C25 to the chassis. But it will not solve your meter/calibration problem, I don't think.

When you said on the previous quote "it's very quiet" it really shouldn't be that quiet.

Did you try SAGE suggestion of using DMM on the O/p and see if you can turn the trimpot for the QBias all the way until the values don't get any higher?
"Now put your DMM across pin 2 and 3 on the output xlr and turn the qbias (in and out knob mid way).  You should see the output moving up or down.  Turn the bias until it gets to the max and doesn't get any higher.  "

Did you manage to do this part?

CHeers

 

[RichardM]

Hi,

First of all , thank you all for your help! I did ground the board. The sound of the problem when the output knob is nearing 6 on the output is similar to that of a clock radio being dialed into it's station. I did use shielded wire for the connection. As for the qbias, I went through the procedure again , using the voltage meter on the output but the q bias doesn't change the output voltage at all when it is being adjusted.


Rich

 

[sage]

I would check the direction of the FETs, it is easy to put one backward and the unit could/would sometimes kinda work but not properly... i went thru that myself... so please check all the FETs and make sure they are facing with their half moon shape the same shape of the printing on the PCB.

worth spending 10 minutes...

sage

 

[jplebre]

I would check the direction of the FETs, it is easy to put one backward and the unit could/would sometimes kinda work but not properly... i went thru that myself... so please check all the FETs and make sure they are facing with their half moon shape the same shape of the printing on the PCB.

worth spending 10 minutes...

sage

and which holes you used (different pinout for 3707/3708 and for 5088

 

[RichardM]

Hi,

Thanks guys! I think I got rid of the output knob problem. Re grounded it and it seems to be fixed. As far as the fets. They are all facing the correct way but they are the 3708s. They are inserted 3 straight across. Are you saying that the center or left(would be left if moon facing the front) should have been inserted in the hole that isn't in line with the others?

Thanks

Rich

 

[sage]

Indeed they DO NOT all face the same way !

just look at the half moon shape  graphic on the PCB from above, now look at the FET from above, like a LEGO game match the shape of the graphic with the shape of the FET. The straight part of the FET goes il line with the straight line, the half moon with the half moon. if you look from above it would make sense.

For example Q11 and Q12 are not facing the same direction while Q13 and Q12 are - same thing Q2 and Q3... not the same...
I think that might be your problem !

sage

 

[RichardM]

Hi Sage,

I should have said that they are seated properly according to their moon shape.

Rich

 

[sage]

then i am sorry but that is as far as my tech help can go - you need someone like Mike at Hairball or more expert techs...

I just had to suggest that 'cause i made that mistake once on Q11 and i had erratic behaviors and could not calibrate... somehow that little silly mistake slowed me down for hours....
I wish I could help more -

 

[RichardM]

Hi Sage,

Thanks for your help. I will eventually figure it out . I hope.

Rich

 

[RichardM]

Hi Guys,

O.k. weird turn of events. I have had issues getting the q bias to work . I was getting fed up but I thought I would try it one more time. I accidentally forgot to click the attack pot to off when I was doing the bias adjustment and found that the q bias worked exactly as it should when the attack pot wasn't clicked to off. My question is: What does this mean? I thought the attack pot needed to be click off to make the bias adjustment? When the attack pot isn't clicked off I can use the q bias to easily make the 1 db drop.

Rich