you need to add a bypass across the NTC, otherwise it will modulate the supply.
inrush current comes from two loads:
a.) start up current for the transformer
b.) charging current for the capacitance after the transformer has started[/QUOTE]..............[/QUOTE]
Does that mean that Panasonic have given up trying to compete in a market from which they can't generate sufficient profit?"Due to declining market demand, Panasonic will discontinue
all Snap-In Aluminum Electrolytic capacitors."
whoa, when did snap-ins become obsolete ????
BTW
Nice amplifier design - and good to see no allergic reactions to use of an output cap
but does charge-up of cap make turn-on thump ? and turn-off thump ?
Yes, they're a large company and have to deal in huge volumes. They're probably being undercut by lower cost producers in China.
A well chosen NTC does not need to be by-passed, the NTC resistance will be low enough. People normally bypass power resistor on a soft start circuit, not NTC.
At 25C the CL-60 NTC resistance is 10ohm and as current goes through the resistance goes down to 0.18ohm if the current through it is 5A.
Measure the current where you want to add your NTC and then choose the right one.
BR,
Eric
No it's not.
read my report of testing the Krell KSA100 Klone.
You need the lowest supply impedance, that the receiver is optimised to use.
All Power Amplifiers, that I am aware of, do not operate as well with added supply impedance.
Do you have a link for your Krell test report ?
Curious to see where your NTC was located in your circuit during your evaluation.
BR,
Eric
the mains had an NTC in the Primary circuit. But it was just at the "testing phase" so no attempt at final assembly.
Unfortunately I did not and still don't have the ability to solve the stability problems . So it is still sitting on the shelf gathering dust. Shame, I had both the 1kVA transformers and 6 big sinks and all the PSUs for a pair of monoblocks.
http://www.diyaudio.com/forums/solid-state/78129-krell-ksa-100mkii-clone-87.html#post1265968
and quite a few posts thereafter.
One thing I do remember is that I could see the effect on the way the signal clipped, or did not clip, as the NTC changed temperature.
Unfortunately I did not and still don't have the ability to solve the stability problems . So it is still sitting on the shelf gathering dust. Shame, I had both the 1kVA transformers and 6 big sinks and all the PSUs for a pair of monoblocks.
http://www.diyaudio.com/forums/solid-state/78129-krell-ksa-100mkii-clone-87.html#post1265968
and quite a few posts thereafter.
One thing I do remember is that I could see the effect on the way the signal clipped, or did not clip, as the NTC changed temperature.
Last edited:
Amp Fine Tuning
Hi,
I'm wondering how to fine tune the amp...I was thinking about the following method :
Add one Source resistor per output device..Rs1, Rs2. I would use 0.33 ohm, 1W
By then varying the value of R3, R5 I would aim to get a current of 0.7A going through Rs1 and Rs2 (about 0.23V across it), for this I would use 2 voltmeters, one per Rs.
My Vgs is very different on each 2SK134 and 2SJ49 so I think fine tuning with the bias current will be better than using the Vgs values.
I do have a scope, can I simply leave R3 as 1K1 and fine tune R5 in order to get symmetrical clipping at the output..??
Once fine tuned I will simply bypass Rs1 and Rs2.
Would this method work ? Hope I'm not to lost..
Thanks for your help.
Eric
Hi,
I'm wondering how to fine tune the amp...I was thinking about the following method :
Add one Source resistor per output device..Rs1, Rs2. I would use 0.33 ohm, 1W
By then varying the value of R3, R5 I would aim to get a current of 0.7A going through Rs1 and Rs2 (about 0.23V across it), for this I would use 2 voltmeters, one per Rs.
My Vgs is very different on each 2SK134 and 2SJ49 so I think fine tuning with the bias current will be better than using the Vgs values.
I do have a scope, can I simply leave R3 as 1K1 and fine tune R5 in order to get symmetrical clipping at the output..??
Once fine tuned I will simply bypass Rs1 and Rs2.
Would this method work ? Hope I'm not to lost..
Thanks for your help.
Eric
Last edited:
This is indeed a shame, instability can be a PITA. If there is no hope with the Krell, can you at least salvage the PS and the HS ?
BR,
Eric
Hi Éric
If you adjust VGS with source resistance present and then after you bypass or remove them then your final bias will be higher then adjusted with the resistors.... This source resistor contributes to reduce the voltage applied to the MOSFET itself thus reduces the bias current.
Fab
Hi Fab,
Thanks, I'm learning
How would you go along tuning an amp with no source resistor ?
It's trickier than I taught...
p.s. What if I use 0.1 ohm instead of 0.33 would it help even though I know that the bias would change slightly once Rs1, Rs2 bypassed.
Each MOSFET have 12Vdc across them, the voltage drop across the source resistor would be about 0.23V if I use a 0.33 ohm. The ratio 12V over 0.23V is very small (not even 2%) but I do understand your concern.
Thanks,
Eric
Thanks, I'm learning
How would you go along tuning an amp with no source resistor ?
It's trickier than I taught...
p.s. What if I use 0.1 ohm instead of 0.33 would it help even though I know that the bias would change slightly once Rs1, Rs2 bypassed.
Each MOSFET have 12Vdc across them, the voltage drop across the source resistor would be about 0.23V if I use a 0.33 ohm. The ratio 12V over 0.23V is very small (not even 2%) but I do understand your concern.
Thanks,
Eric
Last edited:
I'm not trying to match Vgs, I'm aware that they are very different. I simply want to make sure that the amp is properly biased. Sometimes you see amps with a output device heating considerably more than the other..this is what I'm trying to avoid. Maybe this can only happen with a different amp topology.
What would happen if R3 and R5 on the input BJT were very different, ex 1K and 1K5 ?
Thanks,
Eric