dbx 2001 voltage rails (+/-12v to +/- 27v dc!)

[strangeandbouncy]

Hi Guys,

  I came across this dbx2001 document. Apparently, the 2001 will work up to +/-27v rails. I am considering whacking some DOA's for input and output, and was wondering if it would be a good idea to up the supply rails to my VCA's at the same time. CAn someone suggest if there would be any major advantages( apart from headroom?). If I hit the vca's a bit harder, and attenuated the output the same, would I gain anything noise-wise, or would this trade-off be negated by the (possible)increase in distortion? Would I just be shortening the life expectancy of the 2001's, since I guess they would run hotter? If they did run hotter, I guess they might not sound so good?(more errors?)

  Or, . . . . Should I leave well alone, since my curiosity might lead the loss of one of my most important pieces of gear?


    Any comments most gratefully received.


      ANdyP

 

[V9977]

Yes!

I am also on the same boat as you. I am currently working on a possible
design (new pcb) with JLM 99V opamps for i/o, to be run at 24V. I too would
like to use the dbx2001 at elevated voltage with a GSSL type circuit.
Very interested to hear what people have to say!  8)

 

[SSLtech]

thermal noise may increase; and do bear in mind that the 2001's are class-A.

Bearing in mind the scarcity of replacements, I'd be tempted to leave well alone, rather than addressing something which hasn't yet been quantified, assuming that you're looking for a way to elevate above the 2001's anticipated higher noise floor.

For Rock-n-roll, I'd expect any noise benefit to be essentially insignificant; the reliability tradeoff might concern me a little more.

Keith

 

[strangeandbouncy]

Hi Keith,


  Thanks for that straightforward advice. - I'm gonna LEAVE WELL ALONE! As I always say, I am NOT known for quiet mixes anyhow . . .


      Kindest regards,


          ANdyP

 

[V9977]

I on the other hand will apply 24V, heat them up and see where that gets us! 

Thank you for reply SSLtech!
Anyone else?

 

[SSLtech]

I'll keep an eye open for spares to sell to you at a monstrously inflated price! 



Keith

 

[Harpo]

... If I hit the vca's a bit harder, and attenuated the output the same, ...

Aren't these current-in/current-out devices ? You hit them harder by a lower Rin (until they fry). A feedback resistor of same value as Rin at the following inverting op-amp for I/V conversion will keep it at unity for 0V control voltage. Maybe missing something ..
-Harpo

 

[abbey road d enfer]

If I hit the vca's a bit harder, and attenuated the output the same, would I gain anything noise-wise, or would this trade-off be negated by the (possible)increase in distortion?

The input (or output) current is limited by the core current, which is largely dependant on the negative rail voltage. THD increases linearly with (Iin + I out)

Would I just be shortening the life expectancy of the 2001's, since I guess they would run hotter?

  I don't think the increase in temperature would be the most damageable factor; IMO, breakdown voltage is more of a concern.

If they did run hotter, I guess they might not sound so good?(more errors?)

As long as the temperature stays within limits that guarantee a stable operating point, the "sound" shouldn't change.

 

[V9977]

I'll keep an eye open for spares to sell to you at a monstrously inflated price!

;D ;D ;D  !

 

[CJ]

does this apply to the THAT chips?
any advantage raising the rails?

plug in a 2001, stick an ammeter in the power supply lead.
jack up the voltage and watch the milliamps.
i bet there is a very minor increase in current draw compared to the voltage increase.

 

[abbey road d enfer]

does this apply to the THAT chips?

On the chips,the core current is governed by an external res, so if you increase the neg rail, you're supposed to increase this res.

any advantage raising the rails?

Not much latitude there, because the operating voltage range is 4-18V. Not much difference between 12 and 18V; below 12v, headroom diminishes.

plug in a 2001, stick an ammeter in the power supply lead.
jack up the voltage and watch the milliamps.
i bet there is a very minor increase in current draw compared to the voltage increase.

The core current varies almost linearly with supply voltage, but the rest doesn't change as much

 

[V9977]

If the input resistor and output feedback resistor are kept the same as the original
GSSL, would the higher supply voltage (say +/-24V) in the VCA and buffer DOA's allow
for extra headroom in the signal path? A change from 15V to 24V would be 37.5% extra.
That's got to be worth something! Or is the current output the same?  :

Your insight would be much appreciated!

 

[JohnRoberts]

Yes, these VCAs are current ratioing devices, so you do not buy any improvement from upping the voltage rails. The VCA operating current is already a tradeoff between distortion and S/N.

JR

 

[V9977]

:'(

 

[jp-apb]

Others have already noted that the 2001 is current-in/current-out, so increasing the rail voltages doesn't necessarily buy you more headroom.
Take a look at this App Note from THAT Corp that discusses higher rail voltages for VCAs:
http://www.thatcorp.com/datashts/dn123.pdf

The folks that started THAT are the same people that developed the VCAs at dbx.
Look around their website, there's a lot of information available.

JP