TwinLineAmp Opto Compressor TLA2A

[rmaier]

How does this impact the question of metering? Are we back to running a VU meter and trimmer parallel to the 2K series resistor?

Ralph

 

[ruffrecords]

How does this impact the question of metering? Are we back to running a VU meter and trimmer parallel to the 2K series resistor?

Ralph

I think we can retain the 1K8 in series with the 200 ohms and the meter connected across the 200 ohms. All that has changed is the means by which the two channels are linked. We now switch LEDs rathe than coupling the 2 x 200 ohms together.

If we do it like this you can always try out either or both coupling methods.

Cheers

Ian

 

[rmaier]

Thanks, Ian. Parts will be trickling in over the next little while, so I'll start building very soon. I'll leave enough space on my perf board to accommodate a few options for linking.

Ralph.

 

[ruffrecords]

Thanks, Ian. Parts will be trickling in over the next little while, so I'll start building very soon. I'll leave enough space on my perf board to accommodate a few options for linking.

Ralph.

Sounds dangerously like a plan!

Cheers

Ian

 

[pvision]

I think he means this:

Precisely so. Could be done with one DPDT relay if you want to avoid running wires to the switch

Nick Froome

 

[rmaier]

Good idea. I'll give it a go.

 

[ruffrecords]

Good idea. I'll give it a go.

Are you planning to build a stereo compressor in a single box? I find it hard not to think about plug in mixer modules at the moment!!! :-X

I will happily donate a pair of TwinLineAmp boards to the project.

Cheers

Ian

 

[rmaier]

I actually have a pair of TLA v2 boards that I've been sitting on for a couple of years.  Yesterday I started populating them, and I'll be working on putting them in a single box as parts come trickling in. I was pleasantly surprised to get an email from Edcor that the transformers I ordered a few days ago are already in the mail ( I had expected a 4-8 week wait).

So, no need to donate any boards. Your time is already more than enough, Ian.

Thanks!

Ralph

 

[rmaier]

I think we can retain the 1K8 in series with the 200 ohms and the meter connected across the 200 ohms. All that has changed is the means by which the two channels are linked. We now switch LEDs rathe than coupling the 2 x 200 ohms together.

If we do it like this you can always try out either or both coupling methods.

Cheers

Ian

I've stuffed my TLA boards and placed the vactrols and VU circuits on perf board. I don't have any 200r resistors handy... can I get away with 220r or should I wait until I can make another trip to the electronics shop? I spend too much money there, and it's a long drive across town for two resistors!

Ralph

 

[ruffrecords]

I've stuffed my TLA boards and placed the vactrols and VU circuits on perf board. I don't have any 200r resistors handy... can I get away with 220r or should I wait until I can make another trip to the electronics shop? I spend too much money there, and it's a long drive across town for two resistors!

Ralph

220 ohms will be fine.

Cheers

Ian

 

[rmaier]

Thanks, Ian. With luck, I'll be working on wiring things up over the next couple of weeks as parts continue to trickle in - it's exciting to see how things are coming together, and I'm optimistic that I'll have a working unit soon.

I've been wondering about the feasibility of adding a side chain HPF... would it be possible to do something along the same lines as in the Classic Pre (I use the HPF there all the time - it's incredibly useful). Here, could one conceivably insert a rotary switch with several capacitor values just before the threshold pot, or is it more complicated than that? If it's feasible, would you mind pointing me in the right direction for calculating values?

Thanks!
Ralph

 

[ruffrecords]

Yes you could add a HPF just before the threshold pot. A simple 6dB/octave type as in the Classic Solo requires only the addition of a capacitor in series with the input to the threshold pot. The frequency, the pot value and the cap value are related by:

C= 1/(2 x PI x f x R ) Where in this case R=10K

This simplifies to C = 15915/f where C is in nanoFarads. Might as well call it 16000/f

So for 160 Hz you need 100nF. For 80Hz you need 200nF

You can turn it round to calculate the frequency from handy values of capacitance:

f = 16000/C where C again in nanoFarads

So 220nF gives you about 73Hz, 300nF gets you close to 50Hz

These may seem a little inaccurate but rememeber the tolerance on the pot is at least 10% the the frequencies are never going to be exact using this method.

If you have a 2 pole 6 way switch you could cascade a cople of stages to make the ultimate slope 12dB/octave. If you want to go this route let e know and I will sketch something out.

Cheers

Ian

 

[rmaier]

Thanks, Ian, I had hoped as much. My first instinct is to go with a number of series capacitors feeding the threshold pot via a rotary switch resulting in a 6dB/octave slope. It's easy enough to do, and I have the parts at hand. A gentle slope might be a bit less obtrusive, and that would be a plus in terms of the kind of music I usually work on.

On the other hand, there might be situations where a steeper slope would be better... I'm a bit conflicted! Do you have any feelings on which option is most useful?

In the interests of learning a bit more, would you mind sketching out your ideas on implementing a 12dB/octave add-on?

Thanks!
Ralph

 

[ruffrecords]

The very first tube mixer I built had a 2 pole 6 way switch to implement a HPF. I think three positions were 6dB/octave, one was flat and the other two were 12dB/octave. The customer intended to use it for live orchestral recording. I will see if I can find the original schematic.

Cheers

Ian

 

[rmaier]

That could be just the thing...

 

[ruffrecords]

I found the schematic but unfortunately there are no capacitor values on it. I need to did out my old notebook from 10 years ago to find those. However, that old design used 100K pots instead of 10K pots we use today so the values will need to be recalculated anyway.

Cheers

Ian

 

[rmaier]

This looks worth pursuing. I have  some 2 pole 7 position open frame switches that I could try. Hmmm...
How does it work?

Thanks!
Ralph

 

[ruffrecords]

This looks worth pursuing. I have  some 2 pole 7 position open frame switches that I could try. Hmmm...
How does it work?

Thanks!
Ralph

Here is a basic schematic that illustrates how it works:

It is basically two CR 6dB/octave high pass filters one after the other giving a 12dB/octave final slope. The source impedance driving the first one is whatever is connected to the input. Its load is the second stage. The source impedance driving the second stage is the output impedance of the first stage. Its load is the 5K resistor.

You generally make the values of R and C in the two stages the same. So C1 equals C6 and so on. The two pole switch changes the value of the capacitor in both stages at once. For a flat position you just use shorts instead of caps. If you want some switch position(s) to be just 6dB/octave then just replace one of the caps with a short.

The design shown uses 5K resistors so the load on the driving circuit will be half this (2K5). Often it is convenient to make the right hand R a level pot so this would more normally be 10K. So you would also change the first stage R to 10K also and halve the value of both capacitors.

If you want the overall load of the HPF to be about 10K then change the first resistor to 20K or 22K and change the pot to 20K or 22K (depending on what you can get).

The corner frequency is given by 1/(2*PI*R*C) if, and only if, the driving source for each stage is zero ohms and the load on each stage is negligible. The driving impedance for the first stage may well be close to zero and the load impedance of the second stage my well be negligible but neither is true where the first stage drives the second as in this design without a buffer. However, all this does is slightly shift the frequency of both filters. The corner frequency of the first tends to get shifted up  because it is more heavily loaded. The corner frequency of the second stage tends to get shifted down because its driving source impedance is higher. The overall effect is just to make the shape of the turnover a little more gentle but the ultimate slope is unchanged.

Normally you pick a value for the resistors depending on other circuit factors and work out the capacitor values for the frequencies required. If the resistor value is in Kohms the frequency is in Hz and the capacitor value is in nF then the capacitor value required is approximately:

C =  150,000/f*R

Cheers

Ian

 

[rmaier]

Neat. So if I'm understanding this correctly (which isn't a given by any means), using a 20K stepped rotary for the threshold and a 20K resistor between the pole and 0v (for a combined value of 40K), I can flip the  the above formula so that:

f = 150,000/C x 40

If so, then I should expect these common value capacitors to  produce the following filters:

.047uf = 79.7 Hz
.033uf = 113 Hz
.027uf = 138 Hz
.022uf = 170 Hz
.01uf = 375 Hz
.0068 = 551 Hz

I'm inclined to use 12dB slopes for the first three values and 6dB slopes for the rest.

Do these numbers look right or am I missing something?

Thanks!
Ralph

 

[ruffrecords]

Close. You just use the (common)  resistor value in the formula so its 20 not 40. Flipping the formula is fine.

Cheers

Ian