I have used 10r in the base feed (base stopper) to the output driver. It typically shows 1.2mV at quiescent condition, That indicates 0.12mA of quiescent base current.
Easy to monitor with a battery powered milli-voltmeter.
When driven hard into an 8r0 test load, I would expect 2 to 3times that for a loaded base current.
Easy to monitor with a battery powered milli-voltmeter.
When driven hard into an 8r0 test load, I would expect 2 to 3times that for a loaded base current.
10 ohm resistors should read 55mV, which should be easy to read. But I've seen some strange output signals running these inputs when loaded with 100R resistors though. I'm not sure how they would work with the heavier 10R load.
The input Jfets named for BF862 but I have used LSK489 and also tested the 1845 transistors at the input.
You should be able to get a close VAS current reading off R20 and R40 with 100R resistors there.
Where are you planning to adjust DC offset?
Where are you planning to adjust DC offset?
You need to test your OPS. It sounds like you have a problem with it. Disconnect the IPS, install a 15k resistor between V+ and PD+ and another 15k resistor between V- and ND-. You should be able to adjust the bias on the outputs. If not you have an issue with the OPS.
EDIT; In case I wasn't clear, I mean for you to test the OPS stand alone.
EDIT; In case I wasn't clear, I mean for you to test the OPS stand alone.
Last edited:
Yes I can test how much current is going through the R20 and R40 resistor I will update you by tomorrow.
I haven`t made any final version of the pcb as im planning for a 4 layer pcb for best noise perf but for testing I made a 2 layer pcb so didnt plan much for the dc offset right now but if its really high value than i can change it at the input stage by just adding a small value resistor for the input transistors/jfets.
Last edited:
Thank you very much its nice tip to test the OPS alone.
Btw at what voltage do you recommend it to run the OPS when testing.
The voltage is not important. Use the normal rail voltage you plan to use. Use ohm's law to determine the best resistor value. For instance, if you are planning to use +-50v rails and we are shooting for 5.5mA you would use a 9k resistor. 10k will give you 5mA which is close enough to see if your OPS is working properly. What looks wrong to me is that you reported that you have no output current and yet the OPS is over bias. That points to the OPS being the issue. Fix that before you make too many changes to the IPS.
Do you have any tip just to check the IPS + VAS stage like just adding 10k resistor between PD+ and ND- will get almost same voltage, just a little less than the positive psu voltage.
I would just use a 10 ohm resistor from PD+ to the negative feed back and the same from ND-. You can measure across each of those. You should see around 55mV across each one of them. That should reveal if you have a real problem.
do you mean by taking out the feedback resistor and disconnect the output stage?
Or just connect to the base of the input transistor?
You asked about testing the IPS by itself. This is how you do it. With the input shorted and no OPS in the picture, hook up the V+ and V- to the PSU. Connect a 10 ohm resistor from PD+ to the NFB pin on the IPS. Then also connect a 10ohm resistor from ND- to the NFB pin. Energize the IPS and measure "across" each of the 10 ohm resistors. You should see in the neighborhood of 55mV. If not report what you do have and we can take if from there.
Here are my observations:
1. Initially when R21 is at 62ohm the voltage measured across the 10 ohm resistor was about 66mv and at R30 ( 100ohm ) it was about 64mv across the which was evident that which is not the right value.
2. Changed the R21 to 100ohm then the V across 10ohm is about 50mv and across R30 its 500mv which was evident that the output current was arround 5ma.
So Im thinking the input+VAS stage is working fine
but are those right values?
Last edited:
5.5mA Vas current is the target for these, but it should operate at 5mA. Do you have DC offset present at the feedback connection with the 10R resistors connected?
Time to test your output boards. With 15K resistors from V+ to PD+ and V- to ND-, do you have control of bias current?
Time to test your output boards. With 15K resistors from V+ to PD+ and V- to ND-, do you have control of bias current?
Yes there was a DC offset at the feedback connection -72mv and after adjusting the one of the resistor at the input transistors to 50ohm it came to -2.8mv keeping the 5ma output current and also 50mv across 10ohm resistor.
Last edited:
Sounds like your input section is working. Now as jwilhelm has suggested it is time to verify the output section is working. Once they work in isolation they are joined and tuned together.
I have another output stage which has single vbe multiplier which i have used it successfully before and I have hooked it to this input stage.
Few observations:
now the output current is about 3.3ma as measured across R30 i dont know why it has dropped to that value.
amp is working properly now bias stable equally both on positive transistor emitter resistor and negative transistor emitter resistor. very happy to see that being stable after long try. Thank you very much jwilhelm, jkuetemann, still4given, Ostripper for helping in getting it up and running.
Few points I would like to improve
One is I feel little bit of high frequency distortion its not large but very very tiny like it doesnt have creamy treble. Bass needs little bit tighter any suggestions would help alot.
current Bias: 60ma per transistor at 25V.
I have a doubt about using 1845 transistor at the input stage. So thinking to use BC550 as it more better at that location would that improve those issues?
Or burning the amp for 100hours might show up some improvement?