I'm trying to get shunts to work properly. For now I didn't connect preamp part of the board. First, when only one channel power supply is working I could get good results with 2x15 V traffo 22VA. Then, connecting another channel power supply I realized that I could only get +-17V on second reservoir caps C11 and C13 with 500mV ripple. I the switched to 2x15V 50VA. Somehow adjusting trimmer R7 changes both output voltage (which is symmetrical btw) and current thru R1 and R8 (5 ohms). If we presume that preamp output consumes 60 mA and the rest of preamp about 10 mA in total about 70 mA per rail. I have set about 150mA (0,75V on R1) and this gives me +-18V output with 2x17V torus and really low voltage across DN2540s less than 3V. I've put 2 x 220 ohm resistors as test for power supplies. Jfets 2n5486 was selected for idss 10,5 mA so R2 and R9 are shorts. How do I get +-15V?
DN2540 CCS current is defined with source resistor only and of course with the mosfet Idss. With 5 ohm you can get about 150 to 200 mA, but I don't think you need more.
The trimmer R7 is for trimming the output voltage and should not be influenced with the CCS current.
30 to 50 VA 2x15 V AC should be good enough for this shunt regulator. If you can't get the output voltage with it you have some other problem.
I never used 2x17 V AC and never have any problem to adjust the output voltage.
From the batch of bc550 and bc560 I've had, I couldn't get better matched pairs ( hfe yes, Ube NO) and consequently my output is -174 mV in mode without feedback on one channel. It took nearly 45s for servo to bring it down to -0,5 mV. In that moment LF411 output is -2,62 V. I'll try other channel.
If you can't get better match that is still OK, just use delay relay at the output.
I think I should change BD transistors because when I power it up base voltage of Q15 and Q16 is about 10mV and then gradually rises until 0.4V to compensate and make 0v output. I'll disconnect 2x22ohm resistors R42 and R43 to check on output stage only. As I remember Ube difference of BC transistors was about 5mV between npn and pnp under the same condition Ic=2mA measured simultaneously. Now I have only some Cdil(some Indian manufacturer) bd transistors which I didn't test at all.
I still have two last GainWire mk2 boards, one PCB rev 1.1 2mm thick and one PCB rev 1.4 1.6mm thick. Both are the same single side layouts, just rev 1.4 has small heat sinks for the output buffer.
If someone is interested PM me, price is 20 EUR per board, plus shipping.
Damir
If someone is interested PM me, price is 20 EUR per board, plus shipping.
Damir
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.