TG1 - advice from any one who has built it...

[Rob Flinn]

hey,
i do have all the documents. but not really an idea on how to go about it. could you maybe elaborate. i do know where everything goes on the boards - i lack the engineering knowledge.

I think you use a pot connected to where it says i/p pot (instead of jumpering the pads) on the i/o board & leave out the hold pot, but as with all Igors projects he doesn't give you it on a plate.  The 3 that I have built have all been EMI versions so I don't know for sure.  However read the attached folder (part of his documentation) & it may convince you that it is better to keep the hold control like the EMI version rather than the Chandler version wth the input pot.

 

[salomonander]

I think you use a pot connected to where it says i/p pot (instead of jumpering the pads) on the i/o board & leave out the hold pot, but as with all Igors projects he doesn't give you it on a plate.  The 3 that I have built have all been EMI versions so I don't know for sure.  However read the attached folder (part of his documentation) & it may convince you that it is better to keep the hold control like the EMI version rather than the Chandler version wth the input pot.

thanks Rob,
actaully i knew about that option. it only works with 1:2 transformers and im using 1:1 - no intend of changing that. i was thinking of a threshold in a 33609 fashion. just threshold with no effect on volume or the vu meter. but i guess that would be its own little project

 

[Rob Flinn]

Maybe you could connect the 1:1 transformer to the op amp input & make the op amp have a 2x gain to get your level right.

Or do away with the input transformer & use an op amp input.  The first one of these I built you could switch betweenvintage Gardeners transformers i/o & op amp i/o.  I found that the characteristic sound was the pretty similar either way.  I think a lot the vibe comes from the electronics.  One version I built had a Neve style o/p transformer with an op amp input, another had op amp in with Jensen output tranformer, they both sounded very nice.  All of these boxes sound good.

 

[salomonander]

Maybe you could connect the 1:1 transformer to the op amp input & make the op amp have a 2x gain to get your level right.

Or do away with the input transformer & use an op amp input.  The first one of these I built you could switch betweenvintage Gardeners transformers i/o & op amp i/o.  I found that the characteristic sound was the pretty similar either way.  I think a lot the vibe comes from the electronics.  One version I built had a Neve style o/p transformer with an op amp input, another had op amp in with Jensen output tranformer, they both sounded very nice.  All of these boxes sound good.

thanks mate,
ill just keep it as it its. my transformers sound super nice and i dont want to mess with that. now three things im wondering about

- im using 1:1 trafos and have the option to switch between trafos and opamp (opa640). now when switching to the ic version the output is unbalanced. could i balance it by modding the second (now empty) opamp slot on the bypass board?

- my opamp signal is slightly louder. plus im feeling as if i loose a little bass. how can i adress that?

- when switching between comp in / hardware bypass i get some clicks - anything i could do about it (like diodes around the relais)?


thanks a lot

 

[Rob Flinn]

- im using 1:1 trafos and have the option to switch between trafos and opamp (opa640). now when switching to the ic version the output is unbalanced. could i balance it by modding the second (now empty) opamp slot on the bypass board?

I think output isn't unbalanced. I think it is ground compensated, which is good.

- my opamp signal is slightly louder. plus im feeling as if i loose a little bass. how can i adress that?

Use some eq before or after TG ?

- when switching between comp in / hardware bypass i get some clicks - anything i could do about it (like diodes around the relais)? 

The bypass board already incorporates back e.m.f diodes for the relay coils.


thanks a lot

 

[salomonander]

hey guys,
i do need help after all. my weired parasitic effects are back - it seems to depend on calibration. please let me try to explain my issue again:

first off, i think i have the wrong meter? https://www.don-audio.com/Sifam-AL20-6-7-Retro-VU-Meter-Set
anyways thats what i use - but the problems happen with the meter detached as well.

so when i feed no signal and have hold on 0, the needle lingers somewhere in the middle - just about where the original emi meter might have had its 0 marking.
when i fully engage hold, the needle goes down - past 0 on the meter and everything goes bollocks (past the 0 marking). the unit still works with good volume inputs. but when i feed the same signal quieter, the compressor seems to get overtriggered by any little peak reaching the threshold. backing off one position on the hold fixes the issue. as does increasing the input signal.
also, when i go to full hold (needle goes past 0) and back off a step - the needle jumps - then settles again. doing the same thing a step further down doesnt cause the needle jump. so everything with full hold and low input causes issues.

now i am able to cure this by playing with aot1, aot3 and aot5. all of them seem to have an effect on my needles position in full hold position. i trimmed aot1 to not allow the needle past 0 in THD and full hold. and aot3 and aot5 making sure the needle does not go past 0 in limit and hold full. but during this calibration the needle also got jumpy around the 0 mark. still - seems to work now that i set positions that are not past 0 on the meter.

but is this normal behaviour? or do i have a ghost in my system and just found a lucky calibration position?

 

[salomonander]

ill just keep talking to myself here.... maybe someone replies after all
now first of all i get the jumpy meter with no in or output connected. its just in the circuit. when i try to get the needle to 0 on full hold it starts jumping up from 0 and back down. i can varie the lenght of the "jump" by the release switch. so its something triggering the sidechain falsely. also my 10v voltage on the hold switch gets jumpy on max hold.

now another issue im having is that i cant adjust the voltage between VT5 and VT6... which might cause the entire problem.
i am capable of adjusting AOT7 therefore matching the basesupply on VT5 & VT6. this gets the collector voltage to almost the same - but AOT 10 shows no effect at all on T_B and T_C.

matching the base volatges via AOT7 actually cures the needle jumping! but why does AOT 10 not allow me to fully match this (resistance from AOT10 to 20v rail moves correctly)?
any ideas? could my initial zener matching be too bad and responsible for all this?

thanks a lot!

another thing im wondering and need to clarify is the set up procedure. step 5 says:

set unity gain. Set HOLD and OUTPUT to 0 and switch the compressor side chain out by selecting THD. If you have the correct 2.5mA FS meter it should read zero, if not adjust the 100K trim pot until it does. Apply 0dBu (0.775V) 1KHz sine wave to input and measure the output on a oscilloscope or DMM. You need to adjust AOT1/AOT2 so that you get 0.775V at the output.

now, with a 5k linear switch and R115 being 5k for my hold - where would my "Hold 0" be as to referred in the set up? to the far left or in the middle position?

 

[Rob Flinn]

set unity gain. Set HOLD and OUTPUT to 0 and switch the compressor side chain out by selecting THD. If you have the correct 2.5mA FS meter it should read zero, if not adjust the 100K trim pot until it does. Apply 0dBu (0.775V) 1KHz sine wave to input and measure the output on a oscilloscope or DMM. You need to adjust AOT1/AOT2 so that you get 0.775V at the output.

now, with a 5k linear switch and R115 being 5k for my hold - where would my "Hold 0" be as to referred in the set up? to the far left or in the middle position?

Think you need to look at the bit that says "switch the sidechain out". To me this means setting the thd control so no compression is happening ?

 

[salomonander]

yes sidechain is out. but since the hold position still determines the output level in thd mode (at least on my unit) its crucial to know if igor referred to 0 as fully ccw or center dent of a hold pot as on the vintage units.... i think he means center of a 10k pot. but then where is this on my 5k switch & 5k resistor setup?

i dont get it. "Apply 0dBu (0.775V) 1KHz sine wave to input and measure the output on a oscilloscope or DMM. You need to adjust AOT1/AOT2 so that you get 0.775V at the output." how would AOT1 have any effect on the output level in THD mode when this grounds the entire sidechain? at least on my unit AOT1 has no effect on output level in THD the hold control does.

 

[Ilya]

how would AOT1 have any effect on the output level in THD mode when this grounds the entire sidechain? at least on my unit AOT1 has no effect on output level in THD the hold control does.

These set the amount of current through zeners. So yes, they affect overall signal level both in THD and compressor modes.

 

[Rob Flinn]

The way I see it is AOT 1 & 2 affect the way VT23 & VT24 behave which the C.V output comes from.  The C.V output is never disconnected from the audio part of the amp.  Therefore it should have an effect on the level.

 

[salomonander]

Thanks Rob,
i have tow question that might lead to something:

- my hold pot actually changes my main +-20V lines. by .05V  is that ok?

- the control voltage at AOT1 acutally reaches negative values (-0.2V) - thats on full hold. whenever it reaches these minus
  values things go bad. when i dial back the hold from a negative (full hold) control voltage the needle jumps. when i dial it back
  from the slightest (almost full hold) positive control voltage the needle dials back normal. are negative control voltages  ok?

-  i just found that i have a BCY71 in VT27 position. the original schematic asks for a BCY34. Igor used a 214 and i thought it
    would be ok to use BCY71. but maybe its not ?

thanks!

 

[Rob Flinn]

Maybe the bcy 71 has a different pin out ?  I thnk Igor uses a BC214C.  The "C" is important because a BC214 without the "C" has a different pin out.

 

[xarolium]

Wow, -35dBFS is a lot! I thought my units were noisy but they're far from that. At unity gain from input to output I'm getting about -85dBFS, which translates to roughly -72dBu.

I'm using Multicomp 1N751A from Farnell. They are dirt cheap and I had my six matched pairs after testing about 30 pieces.

Hi there,

I'm experimenting exactly the same issue. More or less -36dBFS of background white noise with 1N751A Central Semiconductor Zener Diode.
Does anybody experiment other 5.1V zener Diode like BZX55C5V1, 1N5231A, TZX5V1D to get better S/N ratio ?
It seems that Central Semiconductor Zener are very noisy, what about others manufacturers like On semiconductor or Vishay ?

Your help are welcom

 

[fripholm]

Does anybody experiment other 5.1V zener Diode like BZX55C5V1, 1N5231A, TZX5V1D to get better S/N ratio ?

As you've quoted, I was using Multicomp 1N751A from Farnell in both of my units which did not produce excessive noise. But I wanted to see if even this could be improved and have since tried 1N5231B TR from Mouser (Central Semi) in my stereo unit and the noise floor has dropped almost 3dB compared to the 1N751A.

So you should try different brands to see if that helps the noise. At 0,116 Euros for a diode (when you buy 100) you're not losing too much

 

[xarolium]

As you've quoted, I was using Multicomp 1N751A from Farnell in both of my units which did not produce excessive noise. But I wanted to see if even this could be improved and have since tried 1N5231B TR from Mouser (Central Semi) in my stereo unit and the noise floor has dropped almost 3dB compared to the 1N751A.

So you should try different brands to see if that helps the noise. At 0,116 Euros for a diode (when you buy 100) you're not losing too much

Thank you for your reply. I think i will try 1N5231B and BZX55C5V1 from vishay.
Do you know if all those 5V1 Zener diode have the same curve response ? or do you notify a difference response time between 1N751A and 1N5231B ?

 

[fripholm]

The timing of the compression is controlled by the sidechain - the type of diodes shouldn't affect this. The compression knee and/ or distortion behaviour might change a bit from type to type but in my case I didn't notice anything different.

 

[salomonander]

hey guys,
i think my jumpy needle and weired artefacts are gone by replacing VT27 with a BCY34. need to do some more testing but so far the unit behaves a lot better.
still, im nowhere near unity gain with my hold in the middle (went back to a 10k switch) and thd selected.  i might have to try different zener diodes too. at the moment im using silver motorola 1n751a.

 

[Undos]

Hi

I have spare Original Fairchild 1N751A Zeners from my TG Limiter built which I recently put up on the black market. They worked great in my unit, when matched. I haven't noticed any noise or gain problems.
Shoot me a message when interested.

 

[salomonander]

i tried some 1n751 by RND today - matched them perfectly and ended up in a sea of noise. dont use these!!

i still can not set  AOT10 for adjusting the collectors for VT5 and VT6. the trimmer just doesnt do anything. when measuring resistance it does its job fine. i can measure 0-20ohms to the +20v line but the collector voltage doesnt change a bit
AOT7 has an effect on it but nothing on AOT10. any ideas?