I have the amp assembled now, I used 2x72V AC/1.5kW transformer (the voltage on capacitors is +-100V) and 10 o/p devices (i will add another 10 later - if ill drive 4 ohm load to 1kw)..
The output power should be between 500-600 watts (the amp starts clipping at around volts rms (Up=, but right now im testing without load resistor because I skipped the protection circuit - tomorrow ill drive it with 8 ohm resistor and ill see the actual power) into 8 ohms which is A LOT, althrough the sound quality from what im hearing now isnt what it could be..
So what would I have to do to turn this in amp into a more reliable one with option of balanced input?
Is it normal that that kind of amp needs 6V single ended input to reach its full output voltage?
The output power should be between 500-600 watts (the amp starts clipping at around volts rms (Up=, but right now im testing without load resistor because I skipped the protection circuit - tomorrow ill drive it with 8 ohm resistor and ill see the actual power) into 8 ohms which is A LOT, althrough the sound quality from what im hearing now isnt what it could be..
So what would I have to do to turn this in amp into a more reliable one with option of balanced input?
Is it normal that that kind of amp needs 6V single ended input to reach its full output voltage?
"Is it normal that that kind of amp needs 6V single ended input to reach its full output voltage?"
With no load the amplifier will put out close to 70V RMS, 100V Peak. The voltage gain from the schematic looks like 23.4X so you will need about 4.27V Peak at the input to drive it to full output.
With no load the amplifier will put out close to 70V RMS, 100V Peak. The voltage gain from the schematic looks like 23.4X so you will need about 4.27V Peak at the input to drive it to full output.
Actually, the gain is set by the (R12) 22k/470 Ohm (R11) which is about gain of 46.8 +1 =47.8. This means the amplifier when operating correctly will need about 1.4 volts RMS on the i/p for 70 v RMS on the o/p. = the 1225 Watts RMS into 4 Ohms.
I think there maybe something wrong with this amplifier if you have to put 6 volts into it, to get clipping.
Check the resistor values at the ones mentioned above...
Anthony Holton
I think there maybe something wrong with this amplifier if you have to put 6 volts into it, to get clipping.
Check the resistor values at the ones mentioned above...
Anthony Holton
Thanks for replies
I checked the resistors Anthony mented and they are right values.. I measured again with better voltmeter (my scope isnt calibrated and I can use it only to see the signal, thats why im using voltmeter..) and with 60hz sinus applied, 2.5v rms on input was needed to get to clipping (the amp clips at around 65v rms), maybe something is wrong somwhere else..
Its working great for now, but I didnt drive it to full power yet because I didnt completed protection circuit, but i listened it to 250w on my speakers for half an hour and I could easily hear the change in sound when the bipolars started to work (somewhere at 50w I guess..), the heatsinks were almost cold..
And i measured the bias current of the mosfet on idle it is 3 mA and theres no bias at all on output transistors
, althrough I see no obvious crossover distortion on scope..
And the question again is what would I have to do to turn this in amp into a more reliable one (with standard output stage, where each transistor would have its own bias current..) and with option of balanced input?
I checked the resistors Anthony mented and they are right values.. I measured again with better voltmeter (my scope isnt calibrated and I can use it only to see the signal, thats why im using voltmeter..) and with 60hz sinus applied, 2.5v rms on input was needed to get to clipping (the amp clips at around 65v rms), maybe something is wrong somwhere else..
Its working great for now, but I didnt drive it to full power yet because I didnt completed protection circuit, but i listened it to 250w on my speakers for half an hour and I could easily hear the change in sound when the bipolars started to work (somewhere at 50w I guess..), the heatsinks were almost cold..
And i measured the bias current of the mosfet on idle it is 3 mA and theres no bias at all on output transistors
, althrough I see no obvious crossover distortion on scope..And the question again is what would I have to do to turn this in amp into a more reliable one (with standard output stage, where each transistor would have its own bias current..) and with option of balanced input?
The amp putts out around 420W/8ohm right before clipping with only 4700uF caps on each rail and psu voltage collapsed for 15 volts because of that. Protection shutted down the amp when i putted sine wave in serious clipping and i got 80 v rms on resistor on output of the amp..
The heatsinks were little warm, which surprises me, i thought theyll be hot so i cannot touch them but they are only a little warm, i guess i used too many heatsinks
Which resistors do i have to change then?
The heatsinks were little warm, which surprises me, i thought theyll be hot so i cannot touch them but they are only a little warm, i guess i used too many heatsinks
Which resistors do i have to change then?
Is it normal that R119 (7.5ohm resistor in series with capacitor on output) becomes wery hot and starts to smoke when i apply 20khz sine wave, this starts at 100w and more output power..
I used 1w, maybe i should use more powerful one, can it be ceramic wirewounded or thats not good because of inductance?
I used 1w, maybe i should use more powerful one, can it be ceramic wirewounded or thats not good because of inductance?
Smokin'
Skaara,
You can calculate the power that the 7.5 ohm resistor will dissipate. To do this you first calculate the RMS current flowing through it during a half-cycle, I, and then the average power is P = I^2 x R.
The reactance of the 220nF capacitor is given by Xc = 1/(2.pi.f.C)
where f is the frequency in Hz and C is the capacitance in farads.
At 20kHz Xc = 1/(2 x pi x 3.14 x 20000 x 220e-9) = 36.2 ohms.
The combined series impedance of the 7.5 ohm resistor and the 220nF capacitor is their geometric mean:
Z = sqrt( R^2 + Xc^2) = sqrt( 7.5^2 + 36.2^2) = 36.9 ohms
The RMS current will be the RMS voltage at the output divided by the impedance Z. If your output voltage is 80Vrms then I = 80/36.9 = 2.17amps and the resistor will dissipate P=(2.17)^2x7.5 = 35W.
This is a lot of heat!
You really want to use a low inductance resistor here. You can buy non-inductively wound wirw-wound resistors. Also, note the power is very sensitive to the value of the capacitor. If you used a 100nF then at 80Vrms the resistor will only dissipate 7.5W. You just need to check the amp is stable with the lower capacitor value...in my experience it should be ok.
Skaara,
You can calculate the power that the 7.5 ohm resistor will dissipate. To do this you first calculate the RMS current flowing through it during a half-cycle, I, and then the average power is P = I^2 x R.
The reactance of the 220nF capacitor is given by Xc = 1/(2.pi.f.C)
where f is the frequency in Hz and C is the capacitance in farads.
At 20kHz Xc = 1/(2 x pi x 3.14 x 20000 x 220e-9) = 36.2 ohms.
The combined series impedance of the 7.5 ohm resistor and the 220nF capacitor is their geometric mean:
Z = sqrt( R^2 + Xc^2) = sqrt( 7.5^2 + 36.2^2) = 36.9 ohms
The RMS current will be the RMS voltage at the output divided by the impedance Z. If your output voltage is 80Vrms then I = 80/36.9 = 2.17amps and the resistor will dissipate P=(2.17)^2x7.5 = 35W.
This is a lot of heat!
You really want to use a low inductance resistor here. You can buy non-inductively wound wirw-wound resistors. Also, note the power is very sensitive to the value of the capacitor. If you used a 100nF then at 80Vrms the resistor will only dissipate 7.5W. You just need to check the amp is stable with the lower capacitor value...in my experience it should be ok.
R1 change to 1k and R2 leave at 27k
I would suggest you replace C102 to 100nf and R119 to 10 Ohm 5 watt.
It doesn't matter what type wirewound or otherwise.
This provides a cut off frequency of about 160khz. It presently cutting off at about 95khz with the values shown on the schematic.
C102 however needs to be a mains rated cap x2 type 250V~.
Anthony
I would suggest you replace C102 to 100nf and R119 to 10 Ohm 5 watt.
It doesn't matter what type wirewound or otherwise.
This provides a cut off frequency of about 160khz. It presently cutting off at about 95khz with the values shown on the schematic.
C102 however needs to be a mains rated cap x2 type 250V~.
Anthony
Thank you for info, i get the 275v ac / 100nf MPX x2 capacitor and and 10 ohm / 5w non inductive resistor, I installed these components but i wont test them at 20khz again
when playing music at full power they shouldnt be hot, because high frequencies appear for some short time only.. am i thinking right?
Does this amp have a protection for shorted output and if it does, is it reliable? (at 15v rms 1khz sine wave on o/p i shorted the outputs, the voltage on output of the amp falled for some volts, the relay didnt switched off the output - maybe the protection for shorted outputs is elsewhere in the circuit..?)
Thank you
when playing music at full power they shouldnt be hot, because high frequencies appear for some short time only.. am i thinking right?
Does this amp have a protection for shorted output and if it does, is it reliable? (at 15v rms 1khz sine wave on o/p i shorted the outputs, the voltage on output of the amp falled for some volts, the relay didnt switched off the output - maybe the protection for shorted outputs is elsewhere in the circuit..?)
Thank you
hmm, I tested it with 20khz sine wave in 8 ohm resistor to see if resistor gets hot and I discovered that from about 100-200 watts the picture on scope started to fuzz, and then go back to normal pic of sine wave, then it did it a few times and i lowered the power, maybe it was oscilating? If its so can i use the 220nF cap with this 10ohm resistor, or will i have to buy 7.5ohm one? How important is the right resistance of this resistor?
in 8 ohm resistor to see if resistor gets hot and I discovered that from about 100-200 watts the picture on scope started to fuzz, and then go back to normal pic of sine wave, then it did it a few times and i lowered the power, maybe it was oscilating? If its so can i use the 220nF cap with this 10ohm resistor, or will i have to buy 7.5ohm one? How important is the right resistance of this resistor?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.