Tube compressor output attenuator

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

beatnik

Well-known member
Joined
Oct 18, 2009
Messages
1,179
Location
Italy
I have been experimenting with vari mu tube compressors lately building some classic circuits by urei, rca, etc.
Most of these compressors used a 600ohm T-pad after the output transformer, in the form of massive Daven stepped attenuators which are not so easy to obtain.

There are some 600ohm T-pad attenuators available on the market, but these are in fact standard 3-gang pots in the 1/4W, 1/2W range. I am not sure they can sustain the high levels that can be produced by tube output stages. At certain settings they might just burn out.
I was wondering if the 600ohm T-pad is that necessary after all or it could be effectively substituted by a different attenuator configuration ?


 
Most professional gear, including tube based ones, can output  no more than about +20dBu into 600 ohms which is 100 milliwatts. At Neve back in the 70s we used a simple 2K LIN pot rated at 2W which was slugged by a 1K resistor for this function on balanced monitor outputs. This will give 10dB attenuation t the mid point. There are more complex solutions but this will do in the majority if cases.

Cheers

Ian
 
If I am not mistaken the Davens attenuate up to 40dB, was wondering if this can be achieved with a single section pot ?
 
The mighty HP200cd oscillator uses a special dual pot as output attenuator , its an Allen and Bradley , Im not sure the wattage rating or what kind of resistance curve it would need to function . Might be worth assembling a dual pot with 25k and 100k  log sections and see how it works out .
That unit can push out serious voltages but as far as I can tell its physical wear is the only reason for the pot to stop working. 
Worth a look at the schematic just for reference in any case .
 

Attachments

  • hp-200cd.gif
    hp-200cd.gif
    103.9 KB · Views: 53
beatnik said:
If I am not mistaken the Davens attenuate up to 40dB, was wondering if this can be achieved with a single section pot ?

Do you really need 40dB of attenuation? If the output is +20dBu then with 40dB of attenuation your output drops to -20dBu. Do you want an output trim control or a fader?

It is not too hard to build a simple version of the Daven using a 24 way switch and a bunch of resistors.

Cheers

Ian
 
Tubetec said:
The mighty HP200cd oscillator uses a special dual pot as output attenuator , its an Allen and Bradley , Im not sure the wattage rating or what kind of resistance curve it would need to function . Might be worth assembling a dual pot with 25k and 100k  log sections and see how it works out .

This is a classic bridged T attenuator. At maximum attenuation the junction of the two 620R resistors is connected to cold so the driving output is loaded with 620R and the driven device sees the other 620R as its source. At minimum attenuation the junction of the two 620R resistors should ideally float and the other ends of the 620R resistors are shorted together so the driving output is connected directly to the driven input. The bridged T usually assumes that both the driving source and driven impedance are 600 ohms (as indicated on the schematic).  This is the exact schematic used in Painton quadrant faders but with the added advantage that you can calculate and fit the exact resistors required for each step rather than relying on the dubious law of a pot.

Again, it is not too hard to built one of these using a two gang rotary switch.

Cheers

Ian
 
Thanks for the more in depth explanation on that Ian ,
my brain was having a bit of a wrestling match with itself trying to visualise the operation of it .
 
> sustain the high levels that can be produced by tube output stages.

As limiters? Or as over-driven guitar distortion boxes?

A good US limiter would have two 6V6 so 8 Watts or 8,000mW or 39dBm at primary. Take 2dB in OT and 3dB in the usual 3dB build-out pad is still over 2 Watts. BUT the actual goal was to drive line-loss and deliver +10dBm at the transmitter, with <<1% THD. And the whole point of a limiter is to *control its own output*, so it CAN'T be overdriven like Neil's Deluxe. So there is unlikely to be a tenth-Watt (+20dBm) of audio power available. Even if you only play sustained sine-waves.

And in modern studios you can dump most of that in a 1k 10W (to be sure) fixed resistor, with a 5k pot to dial-off a level that does not smoke your ADC.

Not sure? Put a 600r 1W resistor across the limiter output and beat it. At "sane" settings and level (seasoning, not destructing) the resistor won't get warm.
 

Latest posts

Back
Top