THAT4301 compressor

[abbey road d enfer]

Thanks Abbe!
Yes Im sure!
I measured twice!

I still doubt it. How have you measured it? Digital meter?

Ill need value of components!?

Check my earlier post #202, everything's in it.

 

[iprovlek]

Yes digital meter Abbey!

 

[abbey road d enfer]

Yes digital meter Abbey!

Then are you sure you read correctly? Can you compare with a 2.2 ohm resistor?

 

[iprovlek]

Yes Im sure Abbey!

Im 100%  sure, with digital meter

 

[iprovlek]

Abbey I solve problem with Gain reduction 1mA meter!

I use the first design for this purpose!

 

[iprovlek]

Rin's=10k and Rf=100k

This drive 1mA meter without problem!

I tried "voltage follower"(down below) deign but like you said Abbey,its to low to drive 1mA meter!

 

[snaper]

Compressor done!
Thanks for the etch files and build docs Ivan!
Sounds sooo good, and everything worked at the first power up!

 

[hazel]

Here's mine, working like a charm!

 

[Digitalresonance]

Hi all

I just ordered 2 pcb's. This will be my first diy audio processor build.

I recently decided that I want to make my own hardware, I've been producing for almost 20 years and developed a few of my own Vst plugins too but there's nothing quite like having hardware especially if you made it yourself.

I've built a few noise makers and kits over the years so I'm ok with the basics.

Can't wait to hear how it sounds.

 

[Whoops]

Here's mine, working like a charm!

Looks Great, Congratulations

 

[hazel]

Here's mine, working like a charm!

Looks Great, Congratulations

Thanks Whoops,

i've been working on it during the last 2 years but it's not so much complicated project. I'm just lazy and wanted to be tuned to my taste and needs

 

[Diydiy]

Hi

I am new here, but quite an experienced diy-man. I built this THAT4301 compressor, ordered PCB and IC from Diy-tubes. The shecmatic is 1.5.2 (no Pico mod). I got the compressor working, but I have a distortion problem. The RMS mode is OK. But when I switch to peak mode, I get distorted harmonics. Increacsing the timing capacitor to about 47n to 100n I get rid of audible distortion in SOFT KNEE mode, but the distortion is still presend (although not so bad) in HARD KNEE mode. I have checked everything, but I cannot find any reason. So a couple of questions:

1. I have the optional 2.2M resistor connected, but not the optional 1k5? Would this be the reason?
2. I could not find 28k (R11) resistor, so I used 27k4. Reason for distortion?
3. I have tried to adjust trim pot, but that does not help, so is this distortion a default behavior in peak mode with a small timing capacitor or I am just unlucky?

Thanks in advance for any help!

 

 

[hazel]

is this distortion a default behavior in peak mode with a small timing capacitor or I am just

I didn't use diy-tubes PCB but i built a stereo 4301 compressor and found some issues in PEAK mode with small cap timing capacitor paired with high ratios and low threshold (several db's of GR).

RATIO pot is altered when switching between HARD and SOFT KNEE so  a gentle compression in SOFT is smashed  and distorted in HARD KNEE.

I use PEAK mode and HARD KNEE  with audible distortion with a blend control for parallel compression.

When does distortion happens? Always? Only during peaks with massive amounts of compression?

 

[Diydiy]

I use PEAK mode and HARD KNEE  with audible distortion with a blend control for parallel compression.

When does distortion happens? Always? Only during peaks with massive amounts of compression?

Thank you hazel. The distortion is only present with PEAK mode and HARD KNEE causes far more distortion with a small timing cap.  I use my electric guitar to test, and low E string gives most distortion. Yesterday I read THAT note about auto-trim, and there was mentioned, that low frequencies easily cause ripple, so I think this is a trimming issue. I took my signal generator and an old oscilloscope and tried to do something with the trimmer pot. I turned thereshold pot up and ratio pot down and tried to adjust the pot. I the curve on the oscilloscope screen was somewhat jerky all the time, but somehow I got better results after trimming, so only a litte bit of distortion with low frequencies  PEAK-HARD KNEE mode. The timing cap was 100n,  so I don't know what would be the result with a smaller cap.

I have used relatively mild settings (5k ratio pot turned about half way) and also the thereshold pot around halw way.

The big question is, how to get that trim pot properly adjusted, because I think this is a trimming issue.  My IC might also be a monday manufactured 

And one more question; GR meter. I have read so far, that the DC-reading is about 6mV/dB and the reading is linear? That means that a linear led driver should be used for a led based GR-meter?

 

[hazel]

that low frequencies easily cause ripple, so I think this is a trimming issue.
I have used relatively mild settings (5k ratio pot turned about half way) and also the thereshold pot around halw way.

The big question is, how to get that trim pot properly adjusted, because I think this is a trimming issue.  My IC might also be a monday manufactured 

- Low frequencies easily cause ripple, yes and it's worst in PEAK mode, just use any plugin  with minimum attack and release in a bass track and you'll ge distortion

- It's hard to say how much compression you apply just by pots value. Measure voltage with voltmeter and oscilloscope or send audio out from the DAW to compressor, back to DAW and compare both signals so you'll be more precise.

- Adjusting trimpot: i personally used this method: send a 1kHz tone from my DAW to my compressor (NO compression: RAtio 1:1, threshold all way up) and insert compressor output into a new track in my DAW with a spectrum analyzer (http://www.voxengo.com/product/span/). Turn trimpot slowly until harmonics disappear or fade to minimum

 

[Diydiy]

- Low frequencies easily cause ripple, yes and it's worst in PEAK mode, just use any plugin  with minimum attack and release in a bass track and you'll ge distortion

- It's hard to say how much compression you apply just by pots value. Measure voltage with voltmeter and oscilloscope or send audio out from the DAW to compressor, back to DAW and compare both signals so you'll be more precise.

- Adjusting trimpot: i personally used this method: send a 1kHz tone from my DAW to my compressor (NO compression: RAtio 1:1, threshold all way up) and insert compressor output into a new track in my DAW with a spectrum analyzer (http://www.voxengo.com/product/span/). Turn trimpot slowly until harmonics disappear or fade to minimum

Thank you hazel. I have to try harder.  I have a multiturn (right term?) trim pot, and maybe I have not reached the best setting.

I would still like to know, what timing cap value is most commonly used? I think 10n is quite a small. 

 

[hazel]

I have a multiturn (right term?) trim pot, and maybe I have not reached the best setting.

I would still like to know, what timing cap value is most commonly used? I think 10n is quite a small.

I found the sweet spot of the multiturn  symmetry trimpot about half way  but you have to try it for yourself.

Can't remember which cap values i installed, 4 fast positions (small caps), RMS 10uF and AUTO position (with some LED's).  There was a schematic with some mods around here

 

[ungifted]

New version of PCB available. Suits 2U vertically. More info to come.

 

[ungifted]

Updated docs http://diy-tubes.com/image/data/manuals/That4301-v1.6-full.pdf

 

[tmuikku]

hello! great project!

checked out the document out of curiosity. I noticed a drawing that shows wiring to in / out connectors. Pin1 should be used only for shielding even if unbalanced io is used . Please use pin 3 for 0v reference.

I have run into problems with cheap gear which have trs jacks for unbalanced io pin1 (sleeve) connected to reference and ring to nothing.. no audio with balanced cable if the other end has certain type balanced io (transformer for example). i don't know why they do that kind of connection.

Cable shielding carries unwanted error current only. We don't want error current vary audio reference. Better to get rid of it straight from the connector through chassis to mains ground.

just a piece of information for readers