All things G1176 - the new "repost" thread.

[Andrew]

I have no idea what a 1176 is supposed to sound like, so I'm not sure if mine sounds correct. I had it set to 20:1 with the fastest attack and slowest release. It seems to have a sort of harmonic distortion at most settings.

Raw vocal track:
http://members.shaw.ca/chaostheory/raw.mp3

1176 clone vocal track:
http://members.shaw.ca/chaostheory/1176.mp3

Bypass sounds great, as does the unit with the gain reduction FET taken out. I have calibrated the Q-bias, and cannot get it to sound any better by tweaking that trimmer. I have also tried a couple different FETs. Let me know if you guys have any ideas or comments please.

Thanks,
Andrew

 

[gswan]

Have you tried plotting your compression ratios (ie dB in vs dB out) to see what the transfer characteristics are (ie the knee, threshold, ratio etc)? This will give you a good indication that things are working correctly.

"Bypass" is identical to taking the GR FET out. It simply stops the GR section from operating.

What levels were you using into the 1176 in your recording? The compressed version sounds really woolly compared to the original, and there's a background hum as well. Why do you think it's "harmonic distortion"? It sounds more like soft clipping and/or crossover distortion to me. What does the output waveform look like compared to the input when operating like this?

 

[Andrew]

I haven't tried plotting characteristics, what software could I use for that?

The input level was at about -10dB. I'm still working on getting rid of the hum. I don't know why I thought it sounded like harmonic distortion because it definitely seems more like clipping!

The original track is on top:

I've read through a lot of the posts here, and I have no idea what could be causing this. Is it possible that I just pushed it to hard the FET started to distort?

I've checked all transistor voltages, and I'm getting 0v for both the source and the drain. The gate was around -0.75V.

Thanks again for the help,
Andrew

 

[gswan]

I haven't tried plotting characteristics, what software could I use for that?

A spreadsheet. There's a description here:
http://www.axtsystems.com/index.php?view=article&catid=34%3A1176ln&id=54%3A1176lnratios&option=com_content&Itemid=62

The input level was at about -10dB. I'm still working on getting rid of the hum. I don't know why I thought it sounded like harmonic distortion because it definitely seems more like clipping!

Looks like overcompression. Your attack is so fast and the threshold low, so everything is being squashed and there's no dynamics left. Back off the input a bit so you are not continuously compressing and slow the attack a little so some of the leading edge dynamics are still there.

I've read through a lot of the posts here, and I have no idea what could be causing this. Is it possible that I just pushed it to hard the FET started to distort?

No, the FET acts like a voltage-controlled attenuator that shunts the signal to ground. If you push it hard then your signal will be severely reduced.

 

[Andrew]

Looks like overcompression. Your attack is so fast and the threshold low, so everything is being squashed and there's no dynamics left. Back off the input a bit so you are not continuously compressing and slow the attack a little so some of the leading edge dynamics are still there.

Would that explain the soft-clipping sound?

 

[gswan]

Yes, it can do. It's when everything is squashed into square waves by severe limiting.

 

[Andrew]

Okay. From what I've read, I had thought I would be able to push this unit hard without it sounding like that, or that bad anyway?

How important are the X and Y connections to the operation of the GR FET?

Andrew

 

[gswan]

I'm not sure what you expected it to do with the settings you made, it is a compressor after all. To double check, put a sine wave through the unit and look at the output to get a clear view of the distortion.

The X and Y connections are the outputs and are used for metering purposes.

 

[Andrew]

Fixed! Couple of half-blown parts. I'm using a couple op-amps for the line amp and to balance without using a transformer. Worked out really well! Stoked!

Thanks for the help,
Andrew

 

[gemini86]

can you post another sound clip with this opamp setup in place(and working properly)?

thanks!

 

[Andrew]

Sure.

Raw
http://members.shaw.ca/chaostheory/raw.mp3

Fixed
http://members.shaw.ca/chaostheory/1176-fixed.mp3

It was set at 4:1 with medium attack and release. You can hear it working pretty hard on the biggest peaks and not affecting the rest of the vocals too much. I think it's a pretty dynamic vocal track, and I'm assuming it would work better to put a compressor before an 1176 to tame the peaks (didn't track through a compressor), and then the 1176 to smooth everything out.

Somebody correct me if I'm wrong?

Thanks,
Andrew


Edit: I think I have my ratios reversed somehow, so that setting might actually be 20:1.

 

[gswan]

It was set at 4:1 with medium attack and release. You can hear it working pretty hard on the biggest peaks and not affecting the rest of the vocals too much. I think it's a pretty dynamic vocal track, and I'm assuming it would work better to put a compressor before an 1176 to tame the peaks (didn't track through a compressor), and then the 1176 to smooth everything out.

An 1176 is a compressor. That's what it's designed to do, however you need to learn how to use it under different situations since there are several variables that can be changed (input gain/threshold, attack time, release time, compression ratio and makeup gain). I've used it in a variety of configurations (insert and parallel) with different source material and settings. Last use was a trombone solo on a Jazz track and it performed outstandingly.

 

[Andrew]

Do you ever have soft clipping or other distortion problems when it's not set up just right?

 

[gswan]

Do you ever have soft clipping or other distortion problems when it's not set up just right?

Sure, if you have the gain structure incorrectly set or you overcompress everything to the point where it all becomes square waves then you will get varying amounts of distortion. Sometimes this is used as an effect and can sit well in the mix with other dynamic sounds (eg compressed, overdriven guitar with punchy, dynamic drums). It's a matter of knowing what you want to end up with and how to achieve it with the settings on the tools at hand.

 

[glide 1]

Hi, I decided to rewire a fully working rev j board into a smaller enclosure but now i cannot get it to work, no output.

My unit uses an IC on input and a JLM output tranny, rotary switches.

What i have right now:
* voltages from checkplot are ok
* seems to compress based on what VU meter is doing
* input, attack, release, ratio, meter controls work
* VU works

Continuity was double, triple checked from pads to connectors, to XLR, to ratio and meter boards and everything seems ok.

I'm stumped as to why i am not getting any output from the unit.

Can someone please point me in the right direction on where to start checking.

glide 1

 

[gswan]

Inject a signal into the input of the unit and check that it is appearing on the VU meter. Start at the output sockets and look for the signal back through the unit. The point at which it disappears is where you need to fix it. Use a scope if possible.

Also check your wiring on the output sockets, make sure you have the right pins connected on the XLR's (pin 1 is GND, 2 is hot and 3 is cold).

 

[glide 1]

Inject a signal into the input of the unit and check that it is appearing on the VU meter. Start at the output sockets and look for the signal back through the unit. The point at which it disappears is where you need to fix it. Use a scope if possible.

Also check your wiring on the output sockets, make sure you have the right pins connected on the XLR's (pin 1 is GND, 2 is hot and 3 is cold).

Did what you suggested and tracked back from the output xlr. Turns out that i had traces that had lifted at C15 and possibly at pad 17. Just jumpered them and all is well now. Thanks for the help gswan, appreciate it.

glide 1

 

[Andrew]

Still think I'm having some issues with my 1176 build. Does it make sense that playing around with the Qbias trimmer would also change the ratios?

Thanks,
Andrew

 

[gswan]

Still think I'm having some issues with my 1176 build. Does it make sense that playing around with the Qbias trimmer would also change the ratios?

Qbias must be set correctly and carefully or you will be operating your FET at the wrong point on the transfer curve and end up with distortion and incorrect compressor operation.

If you apply too little bias, the FET will be conducting all the time and will be attenuating your input signal needlessly. The compression ratios will reach their limit earlier on and 20:1 probably won't be anywhere near this.

If you apply too much bias then the FET won't reach conduction until your signal is quite large, and will be operating in a non-linear region of the transfer curve. The ratios won't be correct either.

Have a look at some trasnfer curves here to see this:
http://www.axtsystems.com/index.php?view=article&catid=34%3A1176ln&id=46%3A1176lnfets&option=com_content&Itemid=62

 

[Andrew]

Distortion and incorrect compressor operation is definitely how I would describe what I'm hearing, but I'm almost 100% sure that Qbias was set correctly, and my ratios are correct at this setting.

Is it possible that some flux kicking around in the high Z area of Q1 and the 2M resistors could be causing distortion?