[BUILD] 1176LN Rev D DIY

[quinncmusic]

Update,

Today me and Dbonin spent a few solid hours checking through the whole unit via Skype and I think he has found the root of the problem, but we are unsure of how to fix this.

-We checked all voltages and all looks good except R52/53 which seem to be around 4 volts higher than usual.
-We have ruled out a wrongly placed capacitor or resistor
- I ran the 1k signal through the unit and back into pro tools - shows both the upper harmonics, but still no compression!

Lastly we did the test on the ratio board testing the AC and DC with pad 22 and what we found is that not only is the level low but the ratio selection does not affect the voltage at all like it should! We are getting close...

Does anyone know where to go with this? Again, the unit passes signal fine but no compression happening!

 

[dmnieto]

Update,

Today me and Dbonin spent a few solid hours checking through the whole unit via Skype and I think he has found the root of the problem, but we are unsure of how to fix this.

-We checked all voltages and all looks good except R52/53 which seem to be around 4 volts higher than usual.
-We have ruled out a wrongly placed capacitor or resistor
- I ran the 1k signal through the unit and back into pro tools - shows both the upper harmonics, but still no compression!

Lastly we did the test on the ratio board testing the AC and DC with pad 22 and what we found is that not only is the level low but the ratio selection does not affect the voltage at all like it should! We are getting close...

Does anyone know where to go with this? Again, the unit passes signal fine but no compression happening!

Rules of thumb:

1. If you have both harmonics, the FET is conducting, thus you are compressing.
2. If pad 22 is "too low" it could be that your Q-bias is set in such a way that a lot of the signal is being "dumped" to ground. This may mean that your initial Q bias was already conducting.
3. What do you mean with R52/R53 being 4V higher than expected? which points of the resistors you measured? If your rectified DC voltage on the cathode of CR1/CR2 is 4V higher then definitely your Q-bias is messed up.
Please do the following experiment, i am convinced your q bias is wrong.

Set your 1176 as follows:

- Feed 0.775V RMS to input
- Attack: off
- Release: 7
- 20:1
- Output pot: 24

1. Now put the input pot so the +4 meter is showing -3 dBu.
2. Turn the q-bias pot in the direction in which the meter actually moves towards 0 dBu.
3. Continue turning till you see that the +4 meter does not move any more.
If your meter goes to the right so much that it actually pegs the meter then readjust the input pot again so you have -3dBu.
4. At this point you have taken your FET out of conduction, and you can do the q-bias callibration.
5. Just adjust the input pot so you have +1dBu
6. Adjust q bias by turning the trimmer in the opposite direction to steps 2,3 till you see the meter drop to somewhere between 0dBu and -1dBu. (I like mine at -0.5dB)

Tell me how it goes.

 

[dbonin]

Update,

Today me and Dbonin spent a few solid hours checking through the whole unit via Skype and I think he has found the root of the problem, but we are unsure of how to fix this.

-We checked all voltages and all looks good except R52/53 which seem to be around 4 volts higher than usual.
-We have ruled out a wrongly placed capacitor or resistor
- I ran the 1k signal through the unit and back into pro tools - shows both the upper harmonics, but still no compression!

Lastly we did the test on the ratio board testing the AC and DC with pad 22 and what we found is that not only is the level low but the ratio selection does not affect the voltage at all like it should! We are getting close...

Does anyone know where to go with this? Again, the unit passes signal fine but no compression happening!

Rules of thumb:

1. If you have both harmonics, the FET is conducting, thus you are compressing.
2. If pad 22 is "too low" it could be that your Q-bias is set in such a way that a lot of the signal is being "dumped" to ground. This may mean that your initial Q bias was already conducting.
3. What do you mean with R52/R53 being 4V higher than expected? which points of the resistors you measured? If your rectified DC voltage on the cathode of CR1/CR2 is 4V higher then definitely your Q-bias is messed up.
Please do the following experiment, i am convinced your q bias is wrong.

Set your 1176 as follows:

- Feed 0.775V RMS to input
- Attack: off
- Release: 7
- 20:1
- Output pot: 24

1. Now put the input pot so the +4 meter is showing -3 dBu.
2. Turn the q-bias pot in the direction in which the meter actually moves towards 0 dBu.
3. Continue turning till you see that the +4 meter does not move any more.
If your meter goes to the right so much that it actually pegs the meter then readjust the input pot again so you have -3dBu.
4. At this point you have taken your FET out of conduction, and you can do the q-bias callibration.
5. Just adjust the input pot so you have +1dBu
6. Adjust q bias by turning the trimmer in the opposite direction to steps 2,3 till you see the meter drop to somewhere between 0dBu and -1dBu. (I like mine at -0.5dB)

Tell me how it goes.

We did exactly what you are suggesting two times - I'm convinced the qbias is correct based on what we did.
However, there is no voltage drop measured at the output when GR is kicked in (no change across any ratio).
We tested AC voltage drop across the ratios using pad 22 - voltage drop looked normal.
We tested for negative DC voltage on pad 21 and there was NO change as we cycled through the ratios.

This leads me to believe there is a possibly a short to ground somewhere?  We checked component orientation and nothing appears to be reversed. 
R52/R53 voltage changes as you cycle through the ratios, as expected - but being 4 volts higher than my baseline readings on a working rev d it raised a red flag.  Tracing the circuit,  r52 connects directly to the ratio board via pad 21, so no surprise there.
Every transistor test point was bang on, as was every other test point that we checked on the main PCB (except for the high volts on one side of r52).

Being on skype helped us get through a lot of testing but it can't help with visually inspecting for solder bridges or other funky soldering issues.
We also checked for continuity from the ratio board and pots  to the pads on the PCB (pads 18 thru 22) and it seemed normal.

 

[dmnieto]

We did exactly what you are suggesting two times - I'm convinced the qbias is correct based on what we did.
However, there is no voltage drop measured at the output when GR is kicked in (no change across any ratio).
We tested AC voltage drop across the ratios using pad 22 - voltage drop looked normal.
We tested for negative DC voltage on pad 21 and there was NO change as we cycled through the ratios.

This leads me to believe there is a possibly a short to ground somewhere?  We checked component orientation and nothing appears to be reversed. 
R52/R53 voltage changes as you cycle through the ratios, as expected - but being 4 volts higher than my baseline readings on a working rev d it raised a red flag.  Tracing the circuit,  r52 connects directly to the ratio board via pad 21, so no surprise there.
Every transistor test point was bang on, as was every other test point that we checked on the main PCB (except for the high volts on one side of r52).

Being on skype helped us get through a lot of testing but it can't help with visually inspecting for solder bridges or other funky soldering issues.
We also checked for continuity from the ratio board and pots  to the pads on the PCB (pads 18 thru 22) and it seemed normal.

Very weird, having the right voltages on the transistors would tell me that all the resistors in the GR section should be ok. You checked that there was not a cold solder point in r52/r53 to ground correct?

After that I can only imagine that there is some issue with the resistors in the ratio board. Did you check the values?

 

[quinncmusic]

I double checked solder joints on R52 and 53. Looks good. Pulled the ratio board off and measured the resistors, they are all correct.

I plugged the unit back into pro tools and routed some drums through it. It seems like as I randomly scroll through different ratios from time to time compression will kick in! But it is random, most commonly on 8:1 but sometimes it will start on 20 or 4:1. I should add that it is a pretty violent compression and it sounds relatively the same no matter what ratio is selected. Also, even when I hear compression, the GR meter stays put at the very left on the VU and I cannot move the needle even with R75 and/or the 0 set pot. I know this probably sounds quite bizarre...

 

[TreyDay]

Not sure if this is the right place to ask.
But I'd like to buy the hairball audio 1176 kit ($595) and make the compressor, but I'm afraid I can't or get stuck on issues.
It would be my first self build project.

TreyDay

You will do fine - you'll need a good soldering iron, careful attention to detail and patients.  If you are building a rev D, spend some time up front (while you wait for the kit)  to read through this entire thread.  You will learn loads and there are plenty of us online to offer any help we can if you run into issues.  Mike and Mnats are here regularly, so you will be successful!

My two cents:
1.) Read through all the guides several times before even starting the project.  The more you know up front the easier it'll be when you have a bench full of parts staring back at you. 
2.) read this entire thread before starting
3.) build the kit step by step, carefully following the guides (their advise is proven to be best practice, follow it closely)
4.) test when possible before moving on to a new section/area of the kit

My best advise to anyone starting this project (aside from reading EVERYTHING several times)  is to test as you go when ever possible - there are lots of small tests you can do as you assemble the kit, from stuffing the PCB's (main, ratio, meter) to testing wiring continuity that can save you time later if/when problems surface.

Good Luck,  and welcome to DIY!

Thank you I appreciate your 2 cents.

I'm not entirely a noob when it comes to electronics, but It's been a while since I made projects, I've studied electronics back in the days.
But when I need help, I know some one who can help me out.
I already printed all the pages of the hairball 1176 rev D guide. And I've read it like 3 times already, I will start to read this topic too.

The reason I am buying the complete kit is just to be sure I'll get every right component without hesitating what to buy.
Do you have the rev D's schematics for me? Not 100% sure if mine is correct.

Beside that I will also start a u87 clone build, hopefully both with success! (fingerscross)

Thank you for your time!

TreyDay

 

[quinncmusic]

Update, I'm sort of stuck right now and not sure what to do. I took some of the best pictures I could of the whole unit as well as the PCB from two angles. Maybe someone can spot something visually that could be causing my problems? Its worth a shot I guess. Thanks again to all for their time and support so far!

 

[Spadehead]

Hey all, I posted a while back and have checked the thread but still no answer to my issue.

I have built 2 Rev D's and 1 rev A and they all seem to be at unity when both input and output pots are a 9am whereas everywhere I look people seem to get unity at either 12/12 or 10/2?

Doesn't affect me much but it seems like the input is way hot when i run the input and outputs to midday, meaning that sometimes I need the output pot almost all the way off if I want a lot of compression. This is the same issue in all three builds so am I missing something? Calibration works ok and all visual inspections seem about right too.....

Where should unity gain be and what steps should I take?

Any help appreciated....

 

[dbonin]

I posted a couple of video clips today after finishing a rev D:

Qbias tip:  https://www.youtube.com/watch?v=9yECKyklD0o
Tip if you are having trouble zeroing the meter in GR mode:  https://www.youtube.com/watch?v=PScAEEriE-Q

I hope this helps someone!

-Don

 

[dbonin]

Unity Gain and Impedance Matching: Strange Bedfellows
http://www.rane.com/note124.html

 

[Spadehead]

Thanks for the link - interesting point but my problem is that my units seem to be compressing more than they should - almost like there isn't as much input attenuation happening.

Most folks tend to agree on a starting point of 12/12 and driving more for more compression, leaving quite a lot of wiggle room for high levels of compression (dropping the output pot).

I can't get to that level with a 0VU average input level on the console without having the in and out pots at 9am - meaning that if I push the levels they get FAR too loud and useless- raising the noise floor etc - the only option is dropping the input level on the console - not something I should have to do.

It feels like the input needs to attenuate more so that I need to boost the input pot so compensate if that figures? Or am I going crazy??

Thanks in advance....

 

[dbonin]

Thanks for the link - interesting point but my problem is that my units seem to be compressing more than they should - almost like there isn't as much input attenuation happening.

Most folks tend to agree on a starting point of 12/12 and driving more for more compression, leaving quite a lot of wiggle room for high levels of compression (dropping the output pot).

I can't get to that level with a 0VU average input level on the console without having the in and out pots at 9am - meaning that if I push the levels they get FAR too loud and useless- raising the noise floor etc - the only option is dropping the input level on the console - not something I should have to do.

It feels like the input needs to attenuate more so that I need to boost the input pot so compensate if that figures? Or am I going crazy??

Thanks in advance....

I think this is a really great thing to understand and now that you pose the question I would like to know and understand this better myself - if i understand you correctly,  you want to be able to crank the input to get lots of compression WITHOUT clipping or distorting downstream, but you can't do this with a 0VU input into the 1176 - doing so causes you to have either the INPUT or OUTPUT on the 1176 almost OFF in order not to over saturate the next stage in the signal path?

Just for grins, when you measure the input signal you are using going into the 1176 coming from your console, what do you measure on your DMM for AC volts? 

What is the ideal do you think?    So if i feed my units .775 volts (0 dbu) signal, and i set my input/output to 12/12 on the dials (or somewhere near that) I'm looking for something near  .775 volts on the output?  In other words, we're looking for where unity gain occurs, in relation to the position of the I/O knobs?

 

[dmnieto]

Are your pots linear or logarithmic? Also it is important that if you are using the T-attenuator you feed the 0dBu when the input pot is cranked CCW, as the T-pad can have some losses at the input (if you check the XLR in pads and you move the input pot you will see it fluctuate depending on the output impedance of your generator).

 

[dbonin]

Are your pots linear or logarithmic?

Are you asking about the pots on his console?  The hairball output put is a log pot I believe (standard pot hairball provides).  The t-pad is log as well.

 

[Spadehead]

I think this is a really great thing to understand and now that you pose the question I would like to know and understand this better myself - if i understand you correctly,  you want to be able to crank the input to get lots of compression WITHOUT clipping or distorting downstream, but you can't do this with a 0VU input into the 1176 - doing so causes you to have either the INPUT or OUTPUT on the 1176 almost OFF in order not to over saturate the next stage in the signal path?

Just for grins, when you measure the input signal you are using going into the 1176 coming from your console, what do you measure on your DMM for AC volts? 

What is the ideal do you think?    So if i feed my units .775 volts (0 dbu) signal, and i set my input/output to 12/12 on the dials (or somewhere near that) I'm looking for something near  .775 volts on the output?  In other words, we're looking for where unity gain occurs, in relation to the position of the I/O knobs?
[/quote]

Yes this is exactly my problem. It seems that my equivalent of 12/12 on the dials is actually 9/9, meaning that with lots of compression the output pot ends up nearly off so I'm not exactly in the useful range of the dial!!

I will check voltage readings when I'm next in the studio and have the unit on the workbench and get back to you as this is confusing me!

The pots are te standard log alpha pots that came with the Hairball order, yes.

 

[dmnieto]

Ok, let's do some approx math.

The gain of the preamp section is ~26dB the line output has approx 10dB

At a position of 9PM, considering 300degree pots we have approx 60degree radius, so in log -14dB for each of them.

The input transformer has a 5:2 ratio=> -8dB loss.

So at a 9/9 position (considering transformer gain 0)

Output = 0dBu -8dB - 14dB + 26dB - 14dB + 10dB= 0dBu, so yeah 9/9 seems to be unity gain.

Now concerning the gain.

at a 20:1 ratio, the 1176 starts compressing at approx -24dB,

So at a 9/9 position with 0dBu input you have an input level of -22dBu, you are at the edge of compressing.

So your units are fine, and BTW the ouput pot does not change compression.

 

[dbonin]

So your units are fine, and BTW the ouput pot does not change compression.

Nice, thanks dmnieto!

So far I had two people tell me that their rev D's, going head to head with a UI reissue, were too close to call in every respect.  That's good enough for me.
As for the output pot, as I understand his issue, it's not about it altering compression but about compensating for what appears to be an overly hot signal.  I always thought of the output as "make up gain" after compression knocks down the overall level  (ignoring possible bypass mode usage) so I can understand the question if, no matter where the input knobs sits, the output pot must always be kept almost off to avoid distortion.  That's why I was wondering about the actual signal level going in.

 

[Pkj07]

Hello,

I am in the middle of building my first 1176, and am currently on step 3.  I was successful up to the instillation of the meter switch, and spent today wiring the transformer etc.

I took my time, and followed each step carefully, however when it came time to testing the power section, I couldn't get a reading on my multimeter.

I was wondering if anyone might have some suggestions as to why this was, or common causes. 

PS Does the unit have to be plugged in to successfully get a reading?

Thanks in advance…..
Modify message

 

[Echo North]

Hello,

I am in the middle of building my first 1176, and am currently on step 3.  I was successful up to the instillation of the meter switch, and spent today wiring the transformer etc.

I took my time, and followed each step carefully, however when it came time to testing the power section, I couldn't get a reading on my multimeter.

I was wondering if anyone might have some suggestions as to why this was, or common causes. 

PS Does the unit have to be plugged in to successfully get a reading?

Thanks in advance…..
Modify message

Yes you need to have the fuse in and the unit needs to be on.

 

[Pkj07]

Ohh it needs to be plugged into the mains.  Thanks a million.