I finally got my parts. The MJL** from Reichelt are apparently genuine ON parts. I compared them to pics (there was a thread recently about fake mjl3201 and mjl1302) of ON transistors and they match perfectly, even in the slight imperfections.
Hi John,
Your 'scope traces look as if there is oscillation on the input and output waveforms. Just a low level, constant oscillation.
Is it just that hookup, or does it still look like that? It wouldn't surprise me if it was just the 2u2 cap across the output.
-Chris
Your 'scope traces look as if there is oscillation on the input and output waveforms. Just a low level, constant oscillation.
Is it just that hookup, or does it still look like that? It wouldn't surprise me if it was just the 2u2 cap across the output.
-Chris
Chris, i thought the pic was simply "shaken" ? BTW, the lower trace is the other channel without the 2u2 cap load. I think John can clarify...
Ben, so you're soon member of the club ?
John, the price of the new MJLs should be similar to the other MJLs, but they are not yet available to buy.
Mike
Ben, so you're soon member of the club ?
John, the price of the new MJLs should be similar to the other MJLs, but they are not yet available to buy.
Mike
Yes, your'e right Mike, what it seems an oscillation is not, it's
just my shaken picture taken, at the time of pressing the
"trigger" I've shaken the camera, that's all.
No Anatech, what the caps does (and you can see the difference)
is just the ringing, that's all.
Actually you don't see it in the other picture , it was much more stable.
Anyway I noticied that but I was so excited about the Sym that I didn't want to do the picture again.
Sorry about that.
Your'e right again Mike, the bottom trace is the other channel
with no cap, just the 8 ohm load.
just my shaken picture taken, at the time of pressing the
"trigger" I've shaken the camera, that's all.
No Anatech, what the caps does (and you can see the difference)
is just the ringing, that's all.
Actually you don't see it in the other picture , it was much more stable.
Anyway I noticied that but I was so excited about the Sym that I didn't want to do the picture again.
Sorry about that.
Your'e right again Mike, the bottom trace is the other channel
with no cap, just the 8 ohm load.
That vibration was funny....the fat horizontal line...thick horizontal line is also
The camera vibration?....well....this is a very high frequency vibration.
Someone may invite you to be a computer clock oscilator Mr.Mateus....they will say:
- Please!, put your finger here to produce the needed Clock frequency.
Of course you may hold the camera, as you said, was camera vibration, not your own vibration, so the camera will be needed to produce that vibration.....bad camera...better to fix it....hehe
regards,
Carlos
The camera vibration?....well....this is a very high frequency vibration.
Someone may invite you to be a computer clock oscilator Mr.Mateus....they will say:
- Please!, put your finger here to produce the needed Clock frequency.
Of course you may hold the camera, as you said, was camera vibration, not your own vibration, so the camera will be needed to produce that vibration.....bad camera...better to fix it....hehe
regards,
Carlos
Hi MikeB:
Waiting for the PCB from the maker, I did some mental excercise. I figured that to run the Symasym in class A for a 20 watts output (assuming a Sine wave input), it needs to be biased at 393 mA, or for a 36 watts output, 707 mA. Assuming a large enough heatsink is used, what is your thought on biasing the output FETs that high? Do you see any advantage trying to run Symasym in class A?
It is really great. I am learning quite a bit about electronics trying to understand the Symasym schematics. I hope you don't mind my silly questions though.
I coundn't wait for the PCBs and parts to arrive and start building the amp.
Happy Lunar new year from one of your many admirers,
Waiting for the PCB from the maker, I did some mental excercise. I figured that to run the Symasym in class A for a 20 watts output (assuming a Sine wave input), it needs to be biased at 393 mA, or for a 36 watts output, 707 mA. Assuming a large enough heatsink is used, what is your thought on biasing the output FETs that high? Do you see any advantage trying to run Symasym in class A?
It is really great. I am learning quite a bit about electronics trying to understand the Symasym schematics. I hope you don't mind my silly questions though.
I coundn't wait for the PCBs and parts to arrive and start building the amp.
Happy Lunar new year from one of your many admirers,
Hello tttking,
for 0,707A i got :
1) you know: U = R * I
U= 8 * 0,707 A = 5,66 V
2) you know: P = U * I
P = U * I = R * I^2 = 8 * 0,707A^2 = 4 Watt
for 0,707A i got :
1) you know: U = R * I
U= 8 * 0,707 A = 5,66 V
2) you know: P = U * I
P = U * I = R * I^2 = 8 * 0,707A^2 = 4 Watt
tttking said:
Tim, Uwe: For SINUS output current 707mA you need more than 1A bias current. For 20W to 8ohm you need at least 2.3A bias current. All class "A" of course.
Cheers
Cheers
Hi Tim !
For 20watts RMS into 8ohms of ClassA you need much more idle current, voltageswing would be: sqrt(20*8*2) = 17.8v max current: 17.8/8 = 2.24Amperes. This gives 1.12 Amperes (divided by 2 because of push/pull) of idle current.
I think it's no big benefit to use that high idle current in symasym, but i have not tried...
I don't mind your questions... 🙂
Mike
I guess, now Tim is completely confused...
BTW, sqrt = squareroot
For 20watts RMS into 8ohms of ClassA you need much more idle current, voltageswing would be: sqrt(20*8*2) = 17.8v max current: 17.8/8 = 2.24Amperes. This gives 1.12 Amperes (divided by 2 because of push/pull) of idle current.
I think it's no big benefit to use that high idle current in symasym, but i have not tried...
I don't mind your questions... 🙂
Mike
I guess, now Tim is completely confused...
BTW, sqrt = squareroot
If I remember the right PMA tested class A. But decreased bias to 150mA. Because it is optimal for distortion as he sad.
Hi Chuck, UWE and Mike:
OPPS, You all are absolutely right. My mistake. Now I got it. BiasCurrent= (SquareRoot(20/8))/0.707 = SqRoot(2.5)/0.707 =1.58/0.707 =2.235A. Thanks guys.
Hi Mile:
Why does DOZ run in class A? Was it easier to get good result using simpler circuits in class A, at a cost of lower power at the same heat dispensation?
OPPS, You all are absolutely right. My mistake. Now I got it. BiasCurrent= (SquareRoot(20/8))/0.707 = SqRoot(2.5)/0.707 =1.58/0.707 =2.235A. Thanks guys.
Hi Mile:
Why does DOZ run in class A? Was it easier to get good result using simpler circuits in class A, at a cost of lower power at the same heat dispensation?
Hi MikeB:
I apologize. I just noticed that my finger slipped from 'k' to 'l' and addressed you as Mile.
I apologize. I just noticed that my finger slipped from 'k' to 'l' and addressed you as Mile.
tttking said:
Hi Tim, the DOZ or JLH was designed as ClassA. But with modern output devices you can have ClassAB practically free of crossover distortion, also their improved linearity does it not really make necessary to have ClassA for good sonics. You're right, a pure ClassA can be designed much simpler, and this simplicity often leads to superior sonics. The price for ClassA is obviously high heat dissipation.
Mike
MikeB
I was looking at your suggested layout schematic posted to 00940 on 1-12-06 and was curious as to why you are using two separate bridges for the positive and negative supplies. I thought about doing this recently but then decided against it, reasoning that I wouldn't want my ground returns/reference to have diode junctions in series with them.
I was hard wiring these supplies and decided to stick with what I was familiar with, but would be interested in trying it if there is logic to support it.
I have seen this done on a preamp supply ( < 1amp) with no issue that I could hear.
Thanks, Mike
I was looking at your suggested layout schematic posted to 00940 on 1-12-06 and was curious as to why you are using two separate bridges for the positive and negative supplies. I thought about doing this recently but then decided against it, reasoning that I wouldn't want my ground returns/reference to have diode junctions in series with them.
I was hard wiring these supplies and decided to stick with what I was familiar with, but would be interested in trying it if there is logic to support it.
I have seen this done on a preamp supply ( < 1amp) with no issue that I could hear.
Thanks, Mike
MikeB (the original, of course the more MikeBs the merrier): Thanks for the explanation about class A. The PCBs I ordered just arrived. They look very nice.
Sheldon: Is it OK to ship the PCBs to you via priority mail at $5.40, insured for $50?
Terry: Please let me know you still want to pick it up from me in Cerritos or I should mail you your PCBs. My Email address is in my profile.
Sheldon: Is it OK to ship the PCBs to you via priority mail at $5.40, insured for $50?
Terry: Please let me know you still want to pick it up from me in Cerritos or I should mail you your PCBs. My Email address is in my profile.
No Problem. Let me know how you'd like payment(check, Paypal plus charges, or cash). Here or via PM.
Sheldon
Sheldon
Chuck911 said:
150 mA was optimal for class AB. Increasing iddle current to true class A decreased distortion even more, but this time I did not want to build a heater.
Hi !
There is one additional benefit from ClassA operation, the half wave currents to the PSU are reduced and become full wave currents. In symasym part of these currents find their way into the output and show up in measurings as even harmonics. As these are below -70db they should do no harm. symasym5_3 has less of these than sym5_2 but is not free of them. It's possible that these create the sonics of symasym...
I guess that these are created by some asymetric psrr, i can remove that, but there is the risk symasym will loose it's sonics.
Mike
There is one additional benefit from ClassA operation, the half wave currents to the PSU are reduced and become full wave currents. In symasym part of these currents find their way into the output and show up in measurings as even harmonics. As these are below -70db they should do no harm. symasym5_3 has less of these than sym5_2 but is not free of them. It's possible that these create the sonics of symasym...
I guess that these are created by some asymetric psrr, i can remove that, but there is the risk symasym will loose it's sonics.
Mike