Preamp concept

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Pavel !

I hope your work is going well!


I have problem with the variable resistors you said:
The small trimpots (R9 and R25, or R2 in "values.gif") are used for fine setting of gain - to match gain of both channels. They work as variable resistors, you do not have to care how to connect the middle pin - left or right end, does not matter.

But! Which is the middle one ? On your schematics middle one is 3, but according to most VR of this type(helipots) the middle one schould be 2?

Sorry for stupid questions but I lack profesional knowledge (still learning):rolleyes:
 
Hello,

I changed the circuit a little to my own taste.

Comments?

Best Regards:
Nexus :)
 

Attachments

  • preamp.jpg
    preamp.jpg
    43.4 KB · Views: 1,335
Hi,
I just plug boarded a CCS + Shunt regulator as from PMA's schematic.
I have a problem.

The CCS holds constant current on it's own
3V to 31V input results in 31.60ma upto 31.63mA. That's a variation of just under 0.1% with input voltage.
I would say that is a brilliant result for two components (lm317+39r3).
Adding on the shunt, I set the output to 8.2V feeding 820r load resistor resulting in an output of 10mA.
The 317 was set to 12.5mA (1.241V & R=99r5)
All looked good from input =11.1v upto to 20.1V
at input =25V things have gone wrong. The output still holds steady @ 8.2V but the 317 is not holding 1.24Vacross out/adj.
tried 30V input and the output has now fallen to 8V and the out/adj diff has gone back up to 1.23V.
Something weird is happening.

I tried turning down the output voltage to increase the difference between CCS current and output current,
first to 7V then to 5V and the maximum input voltage came down in step with the output.
26V in 8V out
25V in 7V out
23V in 5V out

I am running from a lab supply and the input smoothing is omitted as are any stability caps around the CCS.

I have now increased the CCS current back to 31.6mA and quickly ran through a range of input voltages. The output drops off similar to before when input voltage goes above 20.5V

Tried increasing the load current to 20mA (16.4V into 820r) and the problem gets worse. The shunt is now sinking about 11mA

Any ideas?
a stability interaction between the CCS & Shunt?
 
Well, I tried reducing 10k to 1k, letting 0.6mA into lt431, that hardly affected the result.
I added a 10nF cap across the pot.
This has helped a lot.
ccs=31mA and output=10mA (8.2V into 820r), the input can swing from 11.1V to 31.5V with no problem.
Still not quite right at higher output currents, but better than it was.
Will try swapping the PMA 1uF cap from BE of output transistor to across the pot.
I'll be back.
 
Hi,
success.
But I had to ignore two components in PMA's schematic.
The 10k shown feeding the tl431 and across the BE of the shunt BD140 had to be reduced below 600r, I used 300r resulting in 2mA through the 431.
I moved the 1uF cap from output to shunt transistor base to connect output to TL431 reference.

The circuit then pulls 50mA from the lab supply and feeds 10mA to 40mA into the output and shunts the remainder.

I tried various input voltages from 11V (8V out + 3) to 31.6V and over two output currents 10mA and 20mA.
Tried 15mA & 30mA outputs from 15V input to 31.6V.
Finally 20mA & 40mA from 20V to 31.6V

I even tried 24.5V output with 28Vinput to 63V input, lm317 max differential=38.5V.

All these tests ran with correct shunt regulation.

Next stage is to look at scope of the output for noise and oscillation. I'll keep my fingers crossed and even fit some sinks to the too hot to touch transistor/reg.
 
www.hifisonix.com
Joined 2003
Paid Member
PMA, the bias resistors are quite low here at 10k. how does this circuit handle being fed from a 'normal' value volume control pot of c. 50k? Do you recommend using a buffer like Grey's 4 JFET balanced version, or maybe something simpler like a straight JFET follower with current source load (a la Borberly for example).

Interesting design by the way - simple with great measured results - how does it compare to other stuff you have tried? I'd be interested to hear this compared to, say. a fully symmetrical JFET based circuit.
 
Bonsai san,

thank you very much for your interest in this circuit.



It is still only simulated, but I will surely build a prototype in a couple of weeks/months. I intend to use blue Alps 2 x 10k, BUT this design will work with 2 x 50k without parameters degradation. BTW, in simulation it looks very, very well, even better, than a very well known preamp. -- Edit, I speak about new version, just missed the thread, sorry.

The old one was open, dynamic, with rugged output stage. I want to keep speed and dynamic properties, output stage, but I want to get very low distortion as well in the new design. The original design cannot handle 50k pot.

Oyasumi nasai (it must be late in Tokyo),

Pavel
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.