Hifisonix kx-Amplifier

Jump to Latest
#786
My heatsink on full class A run at 56 C. On class AB about 33. My heatsinks are a but smaller than I would like.

My room temp is around 22 (winter here but house heated).

I would say for long term reliability, the heatsink temp should not be more than about 60 C.

(I use one of those IR thermometers to measure heatsink temp)

Your heatsinks look quite big Radoslav, so I expect your amp will run a bit cooler.

Second Tony’s comment- nice looking build!
 
Last edited:
#791
I have differences between the measured voltages between Q8 and Q9 emitters on my two amps:
Amp A ± 225mV
Amp B ± 320mV
Class AAB with a supply voltage of ±32.5V DC
Also noted one amp runs around 5°C higher heatsink temperature than the other.

In the 'build stage' I had varied the BC transistor types because of the harmonic oscillation I had been getting.
Amp A is a mix of BC546/BC556/BC557 transistors - all 'B gain' - (±225 mv)
Amp B is based on BC550/BC560 - all 'C Gain' - (±320mV)
All other transistors 'as per design'.
Amps fed separately from the same Ripple Eater (with the ZN wired).

I have checked the component values between the 2 boards and can't find any differences.

Amp A seems closer to the design notes (200 ±50mV) but this was, I presume, at a lower supply voltage?

Would the differences between the 'BC' transistor gains make this difference?

Any other ideas to investigate for this discrepancy in voltages?

Have finally (I hope) managed to sort out the damped oscillation which emanated from the PSU and now adjusting the amps prior to installation. I'll add some notes on the former when I have completed the build.
 
#792
So this is high bias? My first thought is to check if R1 & R18 resistance is the same on both. And that the bias jumper(J8) is working. I think replacing transistors will also change things, especially Q7.
I think Bonsai can give you more details. To me it does not sound like a good idea to have different transistors for L&R channels, better to use the same type of transistors on both.
 
#793
Tony, You will get a spread of c. 20% on bias current, but I’d build the amp with identical parts in all the specific slots. I always use C suffix types, but B and A will also work fine, just stick to the same grades. Rallyfinnen’s advice is good on this point.

When you say distortion was high, how do you mean?
 
Last edited:
#794
Rallyfinnen said:
So this is high bias? My first thought is to check if R1 & R18 resistance is the same on both. And that the bias jumper(J8) is working. I think replacing transistors will also change things, especially Q7.
I think Bonsai can give you more details. To me it does not sound like a good idea to have different transistors for L&R channels, better to use the same type of transistors on both.
Click to expand...

Hi Rally,

Thanks for the suggestions

Both amps are class AAB, 'low bias' (no J8 jumper)
R8 is confirmed as 4K7 on both boards.
Will try changing Q7 to a 'C' grade on Amp 'A' as first measure and see what happens.
 
#795
Bonsai said:
Tony, You will get a spread of c. 20% on bias current, but I’d build the amp with identical parts in all the specific slots. I always use C suffix types, but B and A will also work fine, just stick to the same grades. Rallyfinnen’s advice is good on this point.

When you say distortion was high, how do you mean?
Click to expand...

Hi Bonsai,

No distortion, I was referring to the parasitic oscillation that I reported in post 263 in April last year which was eventually solved by adding the collector resistors to Q1 and Q2.
The amps were originally built using the 'C Grade' BC transistors so I played around with 'Amp A' changing the BC transistors to 'B grade' (as per the BOM) to see if I could get rid of the oscillation.
Now that I'm finalising the build (bolting the boards/heatsinks into the chassis) I noticed a difference in heatsink temperature so that's when I checked the bias current and found the difference.
Anyway, I'll first swop out Q7 on the 'A' Amp and see what happens. If I have enough BC 'C grade' transistors I might replace them all.
There was difficulty in getting BC transistors that I was assured were 'originals' from local suppliers so I was never quite sure if they were not the cause of the problem.
Otherwise I have to order in lots of 50 from RS.
 
#796
Tony, just put the same ones in both amps.

The standing current is not adjustable and this was a design choice to keep the amp simple. You will have differences in current between amps due to resistor tolerances eg the emitter degen resistors), bias controller transistor gains and Vbe’s.

(If you cannot get more transistors, let me know what you have in your junk box and you can use those for the bias controller)
 
#797
Hello,
I am building KX. I would like help.
what to connect in the red circles?
thank you
 

Attachments

  • ovakx.jpg
    ovakx.jpg
    703.4 KB · Views: 203
  • ovakx2.jpg
    ovakx2.jpg
    702.9 KB · Views: 201
#799
Hi all,


I may have missed it but I couldn’t find recommended part number for Sam’s ‘Rallyfinnen’ capacitor. Is there a particular type and style that works best? A recommended Mouser part number would be really helpful in the PDF file description so that one doesn’t have to try and find it.


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