mgQ~~60_3.JPG)
We DIYers do things not only because we need, we do because we can!
Sim shows same rail ripple as measured. (20mVac)
Added a ZEN-V6 like regulater on negative rail and measured rail ripple <1mVac at 2.5A on 16Ohm load .
But now rail is only 40V.
Next step is to reconfigure the PSU with pos&neg regulator.
Aim is to reduce PSU ripple noise incl. its harmonics on the amp output to be not audible.
Maybe some other members will post measured rail and output ripple?
you can easy drop 2 of your thermistors and its all good. just disconnect the right ones
Glad I read this! I was about to order two similar to the ones shown in the earlier post.
Any thoughts on this one
New Version Auto Class A Delay Soft Start Power Protection Board | eBay
Or is there one that's better suited?
Thanks.
Rail voltage for best all-around v3
OK, I have read all 292 pages. You guys are so far above my speed.
Let me say how intrepid you all seem to me but I'm in it whole hog here though I' am also inclined to move more slowly on such a big build...
I want to build F5t v3.
I want to stay pretty well inside the smoke and fire zone.
I have a 30 pound, 400 square inch heat spreader I can mount up to 50 smaller heat sinks to and then run a fan if I need - I think I can dissipate pretty good heat ( I have tested this process on a smaller scale with an Aleph J build...it works well)
I have purchased store boards (toecutter layout V2.2) for 4pr outputs (IRPF240-IRPF9240 / channel.
I assume I'll cascode (v3).
I am at a crossroads though about gain, rail voltage and such.
For durability as much as sound what rail voltage seems likely to give me less heartache?
I'm ok with buying a second set of those rare JFETs for the front end if it means ease of use (devices stay in the safe operating parameters and I don't need a preamp) but it seems from the reading that few are "going there".
How do I find the right balance?
I'm not deeply committed to my present speakers (KEF 104/2s) rather I would commit to building the best F5t v3 in an all around sense.
Thoughts?
OK, I have read all 292 pages. You guys are so far above my speed.
Let me say how intrepid you all seem to me but I'm in it whole hog here though I' am also inclined to move more slowly on such a big build...
I want to build F5t v3.
I want to stay pretty well inside the smoke and fire zone.
I have a 30 pound, 400 square inch heat spreader I can mount up to 50 smaller heat sinks to and then run a fan if I need - I think I can dissipate pretty good heat ( I have tested this process on a smaller scale with an Aleph J build...it works well)
I have purchased store boards (toecutter layout V2.2) for 4pr outputs (IRPF240-IRPF9240 / channel.
I assume I'll cascode (v3).
I am at a crossroads though about gain, rail voltage and such.
For durability as much as sound what rail voltage seems likely to give me less heartache?
I'm ok with buying a second set of those rare JFETs for the front end if it means ease of use (devices stay in the safe operating parameters and I don't need a preamp) but it seems from the reading that few are "going there".
How do I find the right balance?
I'm not deeply committed to my present speakers (KEF 104/2s) rather I would commit to building the best F5t v3 in an all around sense.
Thoughts?
The learning is where the smoke is...
A single set of jfets will suffice for a V3 with 4 output pairs.
Cascoding makes sense if you want to use anything higher than 30V rails.
NP suggests without saying so, that he seems to have planned the the V3 for 50v rails. (.6A/device and 240W dissipation) I don't think that this is necessarily a sweet spot, but should be safe. Given your apprehension, keep it 50V or below.
If you're comfortable with your spreader/multiple sink arrangement, by all means go ahead. Be sure to lap the sinks to the spreader - polishing is good, but lapping them with a bit of grit between will make them mate even better. Depending on your rail target this will get huge. I'm more comfortable using large heat sinks and mounting the devices directly.
AudioSan is right - heat sinking is the first and foremost driver for reliability. Set that go from there.
A single set of jfets will suffice for a V3 with 4 output pairs.
Cascoding makes sense if you want to use anything higher than 30V rails.
NP suggests without saying so, that he seems to have planned the the V3 for 50v rails. (.6A/device and 240W dissipation) I don't think that this is necessarily a sweet spot, but should be safe. Given your apprehension, keep it 50V or below.
If you're comfortable with your spreader/multiple sink arrangement, by all means go ahead. Be sure to lap the sinks to the spreader - polishing is good, but lapping them with a bit of grit between will make them mate even better. Depending on your rail target this will get huge. I'm more comfortable using large heat sinks and mounting the devices directly.
AudioSan is right - heat sinking is the first and foremost driver for reliability. Set that go from there.
the *1.414 is useless in a class A amp. there will be more like VAC*1.3.
2*35VAC to 2*40VAC transformer will do the jobb. depending on your mains voltage, and under wich condition the transformer gives out the raited sec voltage.
i would have used 800VA pr ch as minimum. that gives you less heat, more headroom, less voltage sag, and less chance of mechanical transformer hum.
2*35VAC to 2*40VAC transformer will do the jobb. depending on your mains voltage, and under wich condition the transformer gives out the raited sec voltage.
i would have used 800VA pr ch as minimum. that gives you less heat, more headroom, less voltage sag, and less chance of mechanical transformer hum.
Last edited:
I'm searching a couple of soft start modules, what do you think about these?
Power Soft Start v3
or
Power Soft Start v4
I have another question.. which preamp (gain?!) is suggested for the F5T amp?
Power Soft Start v3
or
Power Soft Start v4
I have another question.. which preamp (gain?!) is suggested for the F5T amp?
Have you built a standard F5?
That will teach you a lot. Even to getting the orientation of the multi-turn pots so that same direction rotation gives the same effect at the measuring location.
I ALWAYS insert the adjustment pot so that clockwise turning of the adjustment screw INCREASES the measured effect.
Sometimes that requires a slight modification of the PCB if it only allows for the staggered pin type.
The 3 pin in-line type can be inserted in both orientations and that leads to confusion at the set up stage. Especially if both channels are done differently.
Then there is the effect of tolerances on the jFETs and mosFETs and what may need to be done to get a full range of set up adjustment.
Use the sqrt(2) factor. IT IS ACCURATE.
Then make allowance for diode Vdrops and resistive Vdrops and ripple voltage and the used portion of transformer regulation.
When you do this, the prediction using your PSU model is far more accurate than guessing at 1.3 for a factor.