Revisiting some "old" ideas from 1970's - IPS, OPS

Jump to Latest
E

evanc

Member
Joined 2009
It seams I have a bit of a problem with both amps. I must have had a brain fart when I was working out the bias current measurement before I put the amps in their cases. I'm reading 80mV across a pair of outputs . If I am calculating correctly thats 80 mV across .44ohms for 181mA current. No wonder the amps are getting hot.... I may need to disassemble them and increase the value of the bias trimmer.
 
vzaichenko

vzaichenko

R.I.P.
Joined 2010
Hi Evan,

Yeah, 80mV is too much. Did you check the VAS current in the front-end? Just measure the voltage over one of the emitter degeneration resistors (connected to VAS emitter) and divide by its value.

Anyway, you can adjust the value of not the trimmer itself, but the value of the resistor in series with the trimmer. As you need to decrease the bias - you need to increase the resistor value.

Cheers,
Valery
 
E

evanc

Member
Joined 2009
Valery hello and thanks for chiming in with help. I wish I was more educated on this stuff but...

Referencing the attached schematics what resistors are the VAS emitter degeneration resistors? How much current would be correct?

For the bias adjustment would you be saying that r34(1k) in the OPS is the one to increase?

Thanks,
Evan
 

Attachments

  • DipTrace Schematic - NS-OPS.pdf
    53.8 KB · Views: 133
  • DipTrace Schematic - Modular Vertical VFA.pdf
    46 KB · Views: 125
vzaichenko

vzaichenko

R.I.P.
Joined 2010
No problem ;)

OK, as I see now I used a Lender arrangement here, so the role of those resistors is a bit more complicated than just degeneration - anyway, you need to measure the voltage over R21, R24 (100R) with no signal (Vertical VFA module). I would expect that voltage to be in the range of 0.8-1.2V.

The one you need to increase on NS OPS board is a combined value of R30 || R33. Right now the combined value comes to 370R. For the beginning - just remove R30 (2.7K). This will result in 430R (as only R33 is left in place). If that's still not enough - increase that one gradually to 470R, 510R, 560R, etc. until you can reach the bias low enough (optimal idle current in my tests was 62.5mA per output pair).

Cheers,
Valery
 
E

evanc

Member
Joined 2009
Valery. Thanks again for the support.

Over r 21 and 24 of the ips I measure about 1.2 volts. Removing r30 seems to be doing the trick. The bias climbs slowly over quite a period of time before it levels out but I look to be in the ball park now. I’ll give an update tomorrow evening. :)
 
pinnocchio

pinnocchio

Member
Joined 2010
quebecdiy.forumactif.com/
I've been working on the Tubsumo IPS while I'm waiting for my Toroidy transformer.

Here's some pics

2018-02-03 15.01.40.jpg

2018-02-03 15.01.50.jpg

2018-02-03 15.01.59.jpg


Looks like I'm going to mount Q14 and Q15 from under the board since my socket is flush with PCB and no space under it.

Do
 
F

filenet

Member
Joined 2013
The TubSuMo IPSs arrived for me 4 days ago. For 6.3V heater supply i use a small transformer, nothing fancy. The amplifier biases easy without any problem, just it needs more time at startup for the offset to settle. There is no hum and the distortion is in the 0.002% region.
It plays lovely, you must try this.
Thanks Jeff and Valery
 

Attachments

  • DSC_0175.jpg
    DSC_0175.jpg
    305.4 KB · Views: 359
vzaichenko

vzaichenko

R.I.P.
Joined 2010
filenet said:
The TubSuMo IPSs arrived for me 4 days ago. For 6.3V heater supply i use a small transformer, nothing fancy. The amplifier biases easy without any problem, just it needs more time at startup for the offset to settle. There is no hum and the distortion is in the 0.002% region.
It plays lovely, you must try this.
Thanks Jeff and Valery
Click to expand...

Excellent build - from what I see - looks very much like our reference one :D
Great job :up:
 
pinnocchio

pinnocchio

Member
Joined 2010
quebecdiy.forumactif.com/
evanc said:
"of the amps I’ve built the tubsumo is one of my favorites. I’d be curious to hear that front end with the big NS OPS. Nice work"
Click to expand...

Thanks evanc, now I really want to listen to it more than ever..! :hphones:

vzaichenko said:
"The boards look beautiful, however I have to mention an important point - Q14 and Q15 placeholders are reserved for optional transistors (bjt or jfers). If you use tubes, those placeholders have to be empty."
Click to expand...

I wasn't too sure since the J117 had traces to the tubes triodes pin directly. I'm happy you confirmed this. Just don't know what would be the point of using anything else than tubes here since it is called Tubsumo? :D

Thanks
Do
 
You must log in or register to reply here.
Top Bottom