T4 optical attenuator for teletronix lA-2A ... and how the compressors works

[Analog_Fan]

Hi there,

I'm collecting info on how to build a "Teletronix LA-2A leveling amplifier".

I would like to know how to build a "T4 optical attenuator" cell.
what are the right LDR's ... and what is the electro luminescence material panel.
Does Mouser sell them?
Where can the suitable electro luminescence panel be obtained in Europe?

What kind of VU meter that can be obtained in Europe would be ok to use?

How does this compressor work?
what kind of work does the "6AQ5A" tube do?
How does it detect excess of line level?
Are there schematics around to do this with solid state technology?

 

[JMan]

You might be interested in this.

The 6AQ5 is part of the sidechain. It sends the control voltage to the T4b. If I understand correctly, R37 (limiter response) and R3 (stereo balance adjust) have a role in determining this voltage, so you might want to take a look at that relationship. EDIT: I think R37 has more to do with frequency response, while R3 has a direct impact on the control voltage that the 6AQ5 sees.

 

[Analog_Fan]

You might be interested in this.

The 6AQ5 is part of the sidechain. It sends the control voltage to the T4b. If I understand correctly, R37 (limiter response) and R3 (stereo balance adjust) have a role in determining this voltage, so you might want to take a look at that relationship. EDIT: I think R37 has more to do with frequency response, while R3 has a direct impact on the control voltage that the 6AQ5 sees.

I'm not sure what is happening!

the electro luminescence panel seams to be powered by C11 above the 6AQ5 tube. the signal "leaving the 6AQ5 tube in to R3 pot meter seams to influence the (second) 12AX7A tube via C9, next to the 6AQ5 tube. is the light always on and pulled down?

what is sidechain?

Who knows the electrical properties of the electro luminescence panel?

 

[Analog_Fan]

Would this material be suitable?

EL-Panel 10cm x 10cm EL-Sheet


Specificaties

• size: 10 cm x 10 cm
• Dikte: 0,4 mm - 1,0 mm
• Bedrijfsspanning: 100 V ~ 220 V Optimaal: 120 V
• Bedrijfsfrequentie: 50 Hz ~ 5000 Hz Optimaal: 1500 Hz - 2000 Hz
• Statische capaciteit: 6 nf / m (20 ° C, de RV <80%)
• Helderheid: 30 cd / m2 ~ 126 cd / m2 (bij 120 V, 200 Hz ~ 2000 Hz)
• Vermogen: 108 mw m ~ 1032 mw / m (bij 120 V, 200 Hz ~ 2000 Hz)
• Levensduur: ongeveer 8000 uur
• Bedrijfstemperatuur / vochtigheid: -10 ° C ~ + 60 ° C / RH <90%
• Opslagtemperatuur / vochtigheid: -10 ° C ~ + 60 ° C / RH <65%
  
  seams to run on 100 Volt.

 

[TheJames]

Whoo boy...This topic has been covered extensively over the years.

Short story...Kenetek (referenced above with his pretty extensive collection of information on the LA2a) is where you should go for in-depth understanding of the LA2a and opto cells.

Let's talk compressors....

From a really high-level, you have an input stage, some sort of gain-reduction element, a "sidechain" that drives the gain-reduction element, and then a make-up gain stage/output stage.

The input stage will typically be a line-receiver or transformer, into some sort of buffer amp. From there the signal will typically split. One path feeds the GR element, and the other path feeds the side-chain.

We'll skip the GR/SC for a moment...That's where all the action is.

The make-up/output stage is just that. It's just an amplifier stage after the GR element that boosts the level and drives the outside world.

Now...The gain-reduction element and side-chain...

These can be VCA's, or FETs, or LDRs or whatever else can be used to attenuate a signal. They have behaviors of their own. How fast they react...How long it takes to recover...How linear their action is. So the "character" of a compressor can generally be understood just by understanding what the gain reduction element is. That's not a hard rule, but a generalization.

The side-chain is a set of electronics that converts and conditions the input signal in a way that can appropriately drive the GR element. In the case of a VCA you need very little voltage to feed the control pin, whereas in the case of a opto-cell like a T4, you need a fairly high voltage to make the EL strip illuminate. In the case of a opto-coupler (i.e. Vactrol), you have to drive an LED which is a different low-voltage signal than a VCA. There are other things you can do in the side-chain as well such as removing low-frequencies to not affect them as much, or emphasizing high-frequencies to over-react to HF content.

In the case of a T4 cell in what's traditionally understood as "opto compressors" you have al EL strip that has some characteristics. Specific characteristics will be the wavelength (color) of the light emitted by the strip, how fast it turns on, how slow it turns off, how bright it gets, etc. Additionally, you have the characteristics of the LDR as well. What wavelengths of light is it most sensitive to, how fast does it activate, how long does it take to go back to "normal."

It is precisely the interaction of characteristics of the EL strip and the LDR that give the opto-cell it's particular behavior. This is something that Bill (I think is his name) at Kenetek has spent quite a bit of time measuring. He has built a computer testing system to test the performance of the LDR's/EL strips and makes perhaps the best T4 cell replacements out there.

Unless you have a desire to go into manufacturing, and assuming you wanted a T4 cell to behave EXACTLY like an original Urei T4 cell, it would be best to just purchase the Kenetek T4 opto cells.

If you want to learn more about the classic Urei opto-compressors, just search this forum for the SA-3a, LA-Z-3A, Bloo LA-2a, or anything about T4B cells. There's almost 20 years of people experimenting and building these things here. And that link that JMan provided is a condensed treasure trove of the exact information you need to understand the LA-2A specifically.

 

[JerryPbury]

You know, these things are cheap to make yourself. Anybody remember the Bloo kits? Built 2 of them years ago. Work flawless and quiet. I think I still have the instructions around here to build your own T4.

 

[Bo Deadly]

Last I checked a Kenetek cell was $99 USD? I used one in my LA3A and it works great.

 

[Analog_Fan]

LA3A

Whoo boy...This topic has been covered extensively over the years.

Short story...Kenetek (referenced above with his pretty extensive collection of information on the LA2a) is where you should go for in-depth understanding of the LA2a and opto cells.

Let's talk compressors....

From a really high-level, you have an input stage, some sort of gain-reduction element, a "sidechain" that drives the gain-reduction element, and then a make-up gain stage/output stage.

The input stage will typically be a line-receiver or transformer, into some sort of buffer amp. From there the signal will typically split. One path feeds the GR element, and the other path feeds the side-chain.

We'll skip the GR/SC for a moment...That's where all the action is.

The make-up/output stage is just that. It's just an amplifier stage after the GR element that boosts the level and drives the outside world.

Now...The gain-reduction element and side-chain...

These can be VCA's, or FETs, or LDRs or whatever else can be used to attenuate a signal. They have behaviors of their own. How fast they react...How long it takes to recover...How linear their action is. So the "character" of a compressor can generally be understood just by understanding what the gain reduction element is. That's not a hard rule, but a generalization.

The side-chain is a set of electronics that converts and conditions the input signal in a way that can appropriately drive the GR element. In the case of a VCA you need very little voltage to feed the control pin, whereas in the case of a opto-cell like a T4, you need a fairly high voltage to make the EL strip illuminate. In the case of a opto-coupler (i.e. Vactrol), you have to drive an LED which is a different low-voltage signal than a VCA. There are other things you can do in the side-chain as well such as removing low-frequencies to not affect them as much, or emphasizing high-frequencies to over-react to HF content.

I would like to know more about this.
How do the tubes work?
what's the job of the 6AQ5 and 12AX7 tube in that section.
How does it measure if the incoming audio is exceeding some threshold.
is it like an inverting opamp? albeit more organic being tubes!

I'm thinking of using some multi op-amp setup to determine of the input.
Veritasium says it's possible is his discus about analog computers, what kind of configuration would make it possible to measure if the level is exceeding a certain threshold and use that result to reduce, the excess to reduce. how could you make a opamp measure from a certain amplitude level. differential amplifier?

In the case of a T4 cell in what's traditionally understood as "opto compressors" you have al EL strip that has some characteristics. Specific characteristics will be the wavelength (color) of the light emitted by the strip, how fast it turns on, how slow it turns off, how bright it gets, etc. Additionally, you have the characteristics of the LDR as well. What wavelengths of light is it most sensitive to, how fast does it activate, how long does it take to go back to "normal."

It is precisely the interaction of characteristics of the EL strip and the LDR that give the opto-cell it's particular behavior. This is something that Bill (I think is his name) at Kenetek has spent quite a bit of time measuring. He has built a computer testing system to test the performance of the LDR's/EL strips and makes perhaps the best T4 cell replacements out there.

Unless you have a desire to go into manufacturing, and assuming you wanted a T4 cell to behave EXACTLY like an original Urei T4 cell, it would be best to just purchase the Kenetek T4 opto cells.

If you want to learn more about the classic Urei opto-compressors, just search this forum for the SA-3a, LA-Z-3A, Bloo LA-2a, or anything about T4B cells. There's almost 20 years of people experimenting and building these things here. And that link that JMan provided is a condensed treasure trove of the exact information you need to understand the LA-2A specifically.

building the device from the schematic is not a big deal, unless i can't read a component value, not yet sure what i'm gonna, wires or PCB.
But the difference is that these are not Eurorack modular synthesizer modules on opamp and ota's.
voltage close to 290 Volt after rectifying, i have tektronix 2212 scope with original probe, not sure if i can measure it. max 50 Volt per division. i can't compare with other thing i have build, not sure what to expect and no means to compare.
the VU meter seams to be a critical elements in operation, but i'm not sure what it's requirements and performance are, nor the cost, brands of these things.

i have understood from here, that i need to build a good power supply for the heaters, it should be clean DC. with no noise.
Maybe i use the precision power supply schematic from "RSD Academy" on YouTube.
I have 2 x 15 Volt AC trafo, back to back and rectified it will be like a calculated voltage of 22 Volt, a LM317 or 78xx regulator, the massive overhead would make them very hot.
But not sure what transistor would serve my purpose.

Than i'm considering the Lundahl LL5402 isolation tranformers from the Pultec EQ schematic from Gyraf Audio(DK) schematic.

But i also want a normal Line level audio input! ... to connect my gear directly to it.

i probably buy those from Bill at Kenetek, than i have a reference.

Than i need to know what the voltage rating of the capacitors should be, will be using Wima.

 

[Analog_Fan]

You know, these things are cheap to make yourself. Anybody remember the Bloo kits? Built 2 of them years ago. Work flawless and quiet. I think I still have the instructions around here to build your own T4.

any PDF / documents you can upload?

Thnkx

 

[igs]

I would like to know how to build a "T4 optical attenuator" cell.

You can read an article I wrote a long time ago with the comparison of the old-fashioned cells I was able to buy back then.

 

[Tubetec]

Do you still produce the T4bx IGS ?

 

[TheJames]

LA3A

I would like to know more about this.
How do the tubes work?
what's the job of the 6AQ5 and 12AX7 tube in that section.
How does it measure if the incoming audio is exceeding some threshold.
is it like an inverting opamp? albeit more organic being tubes!

As for "how do the tubes work" there are literally books with hundreds of pages in them discussing tube design and theory. That's WAY beyond the discussion in simple forum post. From a high-level, you apply a control signal to the tubes grid, then the potential between the cathode and anode of the tube allows for an amplification of control signal that's on the grid. Very similar to a transistor.

I'm thinking of using some multi op-amp setup to determine of the input.
Veritasium says it's possible is his discus about analog computers, what kind of configuration would make it possible to measure if the level is exceeding a certain threshold and use that result to reduce, the excess to reduce. how could you make a opamp measure from a certain amplitude level. differential amplifier?

You might want to study how a compressor actually functions. Effectively when the signal entering the compressor is high enough to activate the gain reduction element (as processed by the side-chain circuit), the gain-reduction element starts turning down the signal. The side-chain effects how fast it attacks, what frequencies it focuses on, and long it takes to recover. As stated in my previous response...The actual GR element (FET, VCA, Optocoupler/opto-cell) has a further affect on the actual "shape" of the attack and release characteristics.

As a side topic...If I were to design a "new" analog compressor, I would probably use a VCA as the GR element with a microprocessor based side-chain. The computer-based side-chain would allow you to "emulate" the characteristics of FET's, Opto's etc, and apply that to the control pin of the VCA. You could setup a tube or solid-state input/output amplifier depending on what you wanted for "flavor." There are some other tricks that you would have to work on considering the inherent latency of A/D conversion and CPU number-crunching to maybe get the "speed" that you would want in the actual action of the compressor.

building the device from the schematic is not a big deal, unless i can't read a component value, not yet sure what i'm gonna, wires or PCB.
But the difference is that these are not Eurorack modular synthesizer modules on opamp and ota's.
voltage close to 290 Volt after rectifying, i have tektronix 2212 scope with original probe, not sure if i can measure it. max 50 Volt per division. i can't compare with other thing i have build, not sure what to expect and no means to compare.
the VU meter seams to be a critical elements in operation, but i'm not sure what it's requirements and performance are, nor the cost, brands of these things.

i have understood from here, that i need to build a good power supply for the heaters, it should be clean DC. with no noise.
Maybe i use the precision power supply schematic from "RSD Academy" on YouTube.
I have 2 x 15 Volt AC trafo, back to back and rectified it will be like a calculated voltage of 22 Volt, a LM317 or 78xx regulator, the massive overhead would make them very hot.
But not sure what transistor would serve my purpose.

Than i'm considering the Lundahl LL5402 isolation tranformers from the Pultec EQ schematic from Gyraf Audio(DK) schematic.

But i also want a normal Line level audio input! ... to connect my gear directly to it.

i probably buy those from Bill at Kenetek, than i have a reference.

Than i need to know what the voltage rating of the capacitors should be, will be using Wima.

Sounds like you need to pick a design goal. If you want an LA2a, honestly, buying a Warm Audio or Klark Teknik "clone" might be more cost-effective. If you absolutely want to DIY one, then I would suggest starting by building one direct from the schematic and learning about how it functions. This forum is literally filled with discussions about LA2a's. And as has previously been stated in this thread...The Kenetek collection of "the secrets of the LA2a" download is probably well-worth the cost of admission and would cut down on alot of the meandering that discussions in a forum takes.

 

[TheJames]

Last I checked a Kenetek cell was $99 USD? I used one in my LA3A and it works great.

Same here. I sacrificed a Kenetek T4B for use in my LaZ-3a's. Works great. There's a certain comfort in knowing that Kenetek has tested and verified the response curves of the components to match the originally intended curves from Urei. It behaves as it's supposed to.

 

[Analog_Fan]

Yeah, i figured that aswel, to use a microprocessor ... to actively measure the input signal on 192kHz or so and reduce the VCA if it overshoots 1 Volt line level ... or some algoryth that analyzes the average amplitude.

but if you wanna do it it good, you need a microchip 32 bit MCU, a high "profile" ADC, something like 18 or 24 bit and 0.1% resistors around the opamps, the problem of these "new" devices is the 3v33, you some somehow to need to scale down the input by 50% and center it / offset if with opamps, before entering the ADC. that's also quite a subject.
something for a later date.

I'm following Professor Aaron Lanterman on tube on Tubes.
I was more wondering on how the system it self work to detect excess in amplitude and reduce by that amount.

Yesterday i came across the La-A3 version schematic, witch is solid state and and be simulated, probably.

Thanks.

 

[CJ]

nice page Igor!

the beauty of the LA2a is the sound of the transformers, tubes and T4 module,

you can use octal tubes, 6SL7 for 12AX7a and 6SN7 for 12BH7 and 6V6 for 6AQ5

 

[JerryPbury]

any PDF / documents you can upload?

Thnkx

I found the document but it’s copyrighted and I can not upload it. The pictures aren’t very clear but without the small circuit board and the parts the instructions don’t make much sense. For reference see Igs post above. #10 what I have is essentially a copy of the Universal Audio model.

 

[Analog_Fan]

nice page Igor!

the beauty of the LA2a is the sound of the transformers, tubes and T4 module,

you can use octal tubes, 6SL7 for 12AX7a and 6SN7 for 12BH7 and 6V6 for 6AQ5

View attachment 91999

is the Lundahl LL5402 transformer used in the Pultec E.Q. schematic and sold at don audio suitable?
for in/out

what is the blue thing, looks like capacitor? ( first on the left top row)
what is that other thing / metal with the "ring" on top(second from the right top row).

What kind of resistors did you use.
i use 0.5 watt 1% metal film resistors in my projects, eurorack.

i want mods:
Line level in/out with jacks aside from the standard transformers with XLR jackhousings.
stereo (internal via SPDT switch).

nut sure, not yet viewed if the local shop has these tubes, but the have the "original" ones.
Don audio sells a clone VU meter, but at 185€, are other ones suitable? that's new area for me.
will any work? the cheap one?

I'm gonna design PCB parts, not gonna wire and wire all of it .. but not use headers, solder wire directly to PCB.

2 should fit in 1 rack case.

I'm not sure if i can get these sockets at my local shop.
my local shop has various grades of the same tube, like 12AX7, between 20 or 50€.

i got another, schematic with 2 LDR's on the input, side 2 side on where the original sits at.

 

[igs]

Do you still produce the T4bx IGS ?

Yes, we do.

 

[Analog_Fan]

Made the Eagle footprint for the 6AQ5A tube, the public Tube library from github didn't include that one.

The penthode symbol was present, as there seam to exist a 6BQ5x as well. but the socket was not.

I use 0.5 watt 1% metal film resistor accept where indicated, good or not?

 

[Analog_Fan]

In my project i always use Wima capacitors, accept these work a bi-polar 12V psu.
the voltage rating decides their size.
I still need to make footprint for the trimmer capacitor, once i start searching for it.

on the top side of the tubes, i guess i need a 450V model, but what about those under or south of the tubes (in the drawing) what kind of voltage rating do they need to have? this C3 capacitor is rated for 450V in the schematic, while C4 is just 25V.

what about the resistors, accept for the one indicated that it need to be a 2 watt version, can i use 1% 0.5 watt metal film resistors?

are the 2 1M pot meter, trimmers? the stereo adjust and the other.
BOURNS 3296W