> https://www.flickr.com/photos/821635...7640966440133/
I am missing one picture with the top cover on ......
Patrick
I am missing one picture with the top cover on ......
Patrick
How close is close enough for the power supply rails? I'm in the midst of testing my regulator boards in my second amp, and I'm measuring +16.05 and -15.85 under load for my first board. I did not measure/match my zeners, which maybe I should have. Does this 1.5% difference matter in our application?
Thanks!
Thanks!
I will try to check the transformer output voltages tonight. I guess my main question is, how close do the regulated voltages "really" need to be? Obviously the closer the better, but will a 1% to 1.5% difference between the positive and negative rails substantially effect the performance of the amp?
I have only tested 1 out of my 4 regulator boards, so it will be interesting to see how the others compare when I test them all this weekend.
I have only tested 1 out of my 4 regulator boards, so it will be interesting to see how the others compare when I test them all this weekend.
A few percent of rail error will not make any difference if you are never clipping the output signal.
Where it might be audible and certainly measurable, is whether the clipping is symmetrical in the +ve and -ve halves of the waveform.
We don't listen to music that is clipped, so it doesn't matter if there is a slightly unsymmetrical clipping of the maximum output.
Where it might be audible and certainly measurable, is whether the clipping is symmetrical in the +ve and -ve halves of the waveform.
We don't listen to music that is clipped, so it doesn't matter if there is a slightly unsymmetrical clipping of the maximum output.
So, here are the pictures of my adjustable zener. It's a TL431 with a 1k smd resistor and a 5k potentiometer in series with a 5k resistor. By the books it should be possible to adjust it between 12.5 and 22.5 V.
It works on the bench, but I did not managed to make it work in the circuit. I used instead of the zener, but I get maximum 14 V output. So at the end I used 19v zener to get 16V out with 19v in.
I'd love to understand why my little creation did not work.
D.
It works on the bench, but I did not managed to make it work in the circuit. I used instead of the zener, but I get maximum 14 V output. So at the end I used 19v zener to get 16V out with 19v in.
I'd love to understand why my little creation did not work.
D.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
An externally hosted image should be here but it was not working when we last tested it.
TL431 needs a cap at the output and some minimum load current.
They say 1µF but I usually use 10µF at least.
And you can probably make a SMD version that might fit.
The other way is to use a small BJT to make a Vbe multiplier. No caps required.
This however is not so thermally stable as the TL431.
Patrick
They say 1µF but I usually use 10µF at least.
And you can probably make a SMD version that might fit.
The other way is to use a small BJT to make a Vbe multiplier. No caps required.
This however is not so thermally stable as the TL431.
Patrick
Last week I receive the transformers and I complete the PSU.
Those are the values from the regulators changing the LEDs with 2x1N4148:
Left: In 21.04V/out +16.04V and in 21.05V/out -16.16V
Right: in 21,06V/out +16.12V and in 21.04V/out -16.06V
With the PSU I complete also the bias for the 2 channel (with 4 ohm dummy load but still to be floated):
220.2 mV - 220.8mV with offset -0.4 / +0.5mV
220.3 mV - 220.5mV with offset -0.1 / +0.3mV
220.3 mV - 219.8mV with offset 0 / +0.3mV
220.7 mV - 220.6mV with offset 0 / +0.3mV
I have a dumb question: I notice that the bias is extremely sensible to the temperature. Just breating in the heatsink and the values change. During the bias (at least my setup) the heatsink is not in the final position and the heat remain between the table and the heatsink itself. When will be in the correct position the hot air can flow through the heatsink but the mosfets are also "closed" in the enclosure that, again, is a different condition from the bias.
Most of you have been completed and closed the ampli. So, which is the appropriate value on bias BEFORE to close the amplifier?
I notice also that the offset is moving quite a lot (this is the reason why I type 2 values for the offset). I don't remember the same variations with the F5T V2. Is that normal?
Anyway I am quite happy with the values above. Everything is working as expected.
I still can't float the X... I need a 8 ohm dummy load that is on the way.
If you have any suggestion or advice please let me know
Thanks and Regards,
Enrico
PS: I damage the protection PCB, likely I can get a spare from Mark (thanks Mark). I order a big magnifier and I hope I can win the war with the SMD.
@Ramallo, is really frustating, isn't?
Those are the values from the regulators changing the LEDs with 2x1N4148:
Left: In 21.04V/out +16.04V and in 21.05V/out -16.16V
Right: in 21,06V/out +16.12V and in 21.04V/out -16.06V
With the PSU I complete also the bias for the 2 channel (with 4 ohm dummy load but still to be floated):
220.2 mV - 220.8mV with offset -0.4 / +0.5mV
220.3 mV - 220.5mV with offset -0.1 / +0.3mV
220.3 mV - 219.8mV with offset 0 / +0.3mV
220.7 mV - 220.6mV with offset 0 / +0.3mV
I have a dumb question: I notice that the bias is extremely sensible to the temperature. Just breating in the heatsink and the values change. During the bias (at least my setup) the heatsink is not in the final position and the heat remain between the table and the heatsink itself. When will be in the correct position the hot air can flow through the heatsink but the mosfets are also "closed" in the enclosure that, again, is a different condition from the bias.
Most of you have been completed and closed the ampli. So, which is the appropriate value on bias BEFORE to close the amplifier?
I notice also that the offset is moving quite a lot (this is the reason why I type 2 values for the offset). I don't remember the same variations with the F5T V2. Is that normal?
Anyway I am quite happy with the values above. Everything is working as expected.
I still can't float the X... I need a 8 ohm dummy load that is on the way.
If you have any suggestion or advice please let me know
Thanks and Regards,
Enrico
PS: I damage the protection PCB, likely I can get a spare from Mark (thanks Mark). I order a big magnifier and I hope I can win the war with the SMD.
@Ramallo, is really frustating, isn't?
Wow John, SR009 and BHSE? That's one nice setup. I've got a pair of 009's on order direct from Japan to run with my KGSSHV (might be adding a EC Electra for fun as well).
Great to hear about the F5X working great with headphones. I keep meaning to try out my AKG K1000's out the of the F5X. I bet it will be a great combo. I'll report back when I finally get around to it.
Great to hear about the F5X working great with headphones. I keep meaning to try out my AKG K1000's out the of the F5X. I bet it will be a great combo. I'll report back when I finally get around to it.
Hey Greg, A mate “nattonrice” had his DIY BHSE up for sale, so I snatched it up, great builder I really trust.,the 009’s are a new addition, had them about 5 months now, loving them, sounded terrific with the HV too.
The K1000s should be a GREAT match with the F5X, I have no doubt. I’m loving the F5X as a headphone amp it really is wonderful with any phone I’ve thrown at it.
Congrats with the 009s, can’t go wrong, especially with the HV powering them.
The K1000s should be a GREAT match with the F5X, I have no doubt. I’m loving the F5X as a headphone amp it really is wonderful with any phone I’ve thrown at it.
Congrats with the 009s, can’t go wrong, especially with the HV powering them.
Last edited: