The FH9HVX will be the first to try it. Relatively easy Class AB load. If it works, I might stick either an Alpha Nirvana or a MOFO on it next. For Alpha Nirvana it needs to handle about 105w total. I think it will be fine. Main thing is how well the heat can spread from the thermal pad on the MOSFET to the rough cast iron. Milling it flat would be very helpful.
And if it works, there is a new $10 solution to the very expensive heatsinking problem of hot Class A amps.
Need some guidance please. I am trying to use existing parts I have on hand. I have 4 300VA toroidal transformers. Two transformers have 2 x 22Vac secondaries, and the other two transformers have 2 x 24Vac secondaries.
I could build 2 separate psu's using the transformers 24 V secondaries, but that would only supply about 34V, which is a bit to low for my liking.
Could I wire either sets of the transformers in series, to increase the output voltage and use as a single supply, or would the voltage be to high?.
If the above is possible (and the V is to high), is their a way to reduce the voltage without removing some windings, or should a person just purchase the required transformer.
Thanks for the help,
MM
I could build 2 separate psu's using the transformers 24 V secondaries, but that would only supply about 34V, which is a bit to low for my liking.
Could I wire either sets of the transformers in series, to increase the output voltage and use as a single supply, or would the voltage be to high?.
If the above is possible (and the V is to high), is their a way to reduce the voltage without removing some windings, or should a person just purchase the required transformer.
Thanks for the help,
MM
If you have a transformer for each channel, each will need its own CRC. Although Vunce tested straight bulk caps of about 10,000uF per rail no CRC and that worked the best from standpoint of bass authority. Easy test is to jumper the two 4700uF on CRC with a wire. Listen for yourself and if quiet and no hum then the R is not needed.
Not sure of the exact grade of that cast iron you have, but here's a typical example: MatWeb - The Online Materials Information Resource
vs aluminum
MatWeb - The Online Materials Information Resource
Conductivity of the aluminum is 4.5 x higher than that of the cast iron.
Yes, I know that it is. Just hoping that even as low as it is, it can dissipate say 20w per MOSFET for brief periods. Normal program material is circa 25wrms max. So about 5w per MOSFET playing loud (Assuming 65% efficiency for Class AB).
It’s going to be fine. For 45w per MOSFET like on Class A Alpha Nirvana - probably not.
While waiting for the FH9HVX verification boards to arrive I had plenty of time to play around with a few different power supply options, both SMPS and linear. The SMPS units I tried were a Cresnet SMPS1200-53 and an Abletec ALP0400, both units provided +/-53vdc. I couldn’t get the Crestnet unit quiet enough to be acceptable, there was a ground hum and a background hiss that I couldn’t tame. I tried a few different grounding schemes with no luck. This is strange because I use the exact same SMPS unit with a stereo PeeCeeBee V4H and it’s dead silent?
The Abletec unit worked very nice, no ground hum, very little background hiss. Out of curiosity, I added a CRC filter for each channel between the psu output and amplifier boards (4700uf-1R-4700uF) Much to my surprise, the Abletec powered up just fine with 37600uF added capacitance on the output! Soundwise, I didn’t notice much difference.
Now the linear PSU’s:
First up, a 32V dual secondary trafo fed one of Prasi’s CRC boards populated with 22000-0.22R-22000 and good for 45VDC output. This board was originally built for a Class A amplifier project. The combo worked very well, no background noise at all, dead silent. I used the same CRC board with a lower 28V dual secondary trafo and got 39.5VDC. This also produced pleasant results, all listening was done at reasonable volume level, approximately 80dB at seating position.
Last up was a 34V dual center tapped secondary transformer. Since this trafo is a CT variety I used a different PSU board. The first set-up used one secondary with one PSU board consisting of ten 4700µF bulk caps in a CCCCC arrangement. The combo delivered 47.5VDC. This combo worked very well powering both channels, just a very faint hum with my ear pressed right up to the speaker. I wanted to see how low in capacitance I could go before I noticed a difference, so I whipped up another PSU board with ten 2200µF caps. This board, with only 11000uF per rail I did noticed an increase in hum powering both channels.
Now I have two PSU boards to utilize both center tap secondaries. This combo is essentially a dual mono set up. One channel is fed with 23,500µF per rail the other channel is fed by 11,000µF per rail. There is a very faint hum from the lower capacitance psu, the larger psu is dead quiet.
It’s probably not fair to compare this set up to the previous ones because it’s the only true dual mono set up, the closest was the Abletec with the added CRC networks. There is a noticeable improvement in bass, soundstage and midrange smoothness. Also, The music seems a bit more lively, a better “toe tap” feeling I guess you would call it.
In my very unscientific comparison this linear all C’s combination works best to my ears. I will go with a dual mono PSU with 20-25,000µF per rail for each FH9HVX board.
HAPPY BUILDING ALL!!!
The Abletec unit worked very nice, no ground hum, very little background hiss. Out of curiosity, I added a CRC filter for each channel between the psu output and amplifier boards (4700uf-1R-4700uF) Much to my surprise, the Abletec powered up just fine with 37600uF added capacitance on the output! Soundwise, I didn’t notice much difference.
Now the linear PSU’s:
First up, a 32V dual secondary trafo fed one of Prasi’s CRC boards populated with 22000-0.22R-22000 and good for 45VDC output. This board was originally built for a Class A amplifier project. The combo worked very well, no background noise at all, dead silent. I used the same CRC board with a lower 28V dual secondary trafo and got 39.5VDC. This also produced pleasant results, all listening was done at reasonable volume level, approximately 80dB at seating position.
Last up was a 34V dual center tapped secondary transformer. Since this trafo is a CT variety I used a different PSU board. The first set-up used one secondary with one PSU board consisting of ten 4700µF bulk caps in a CCCCC arrangement. The combo delivered 47.5VDC. This combo worked very well powering both channels, just a very faint hum with my ear pressed right up to the speaker. I wanted to see how low in capacitance I could go before I noticed a difference, so I whipped up another PSU board with ten 2200µF caps. This board, with only 11000uF per rail I did noticed an increase in hum powering both channels.
Now I have two PSU boards to utilize both center tap secondaries. This combo is essentially a dual mono set up. One channel is fed with 23,500µF per rail the other channel is fed by 11,000µF per rail. There is a very faint hum from the lower capacitance psu, the larger psu is dead quiet.
It’s probably not fair to compare this set up to the previous ones because it’s the only true dual mono set up, the closest was the Abletec with the added CRC networks. There is a noticeable improvement in bass, soundstage and midrange smoothness. Also, The music seems a bit more lively, a better “toe tap” feeling I guess you would call it.
In my very unscientific comparison this linear all C’s combination works best to my ears. I will go with a dual mono PSU with 20-25,000µF per rail for each FH9HVX board.
HAPPY BUILDING ALL!!!
Last edited:
Thank you, Vunce! Great write up of your listening tests - very helpful for us to make a choice. What if looks like though is that the FH9FVX is very flexible with different types of PSU’s but has great PSRR that just using lots of bulk capacitance seems to do the trick.
With that, I have perhaps the “All C’s PSU”. If uses all 4700uF caps to get about 20,000uF per rail. There is an optional CCRCC setup by using an R instead of a wire jumper.
If JPS64 has time, maybe he can make a layout of the “All C’s” PSU?
With that, I have perhaps the “All C’s PSU”. If uses all 4700uF caps to get about 20,000uF per rail. There is an optional CCRCC setup by using an R instead of a wire jumper.
If JPS64 has time, maybe he can make a layout of the “All C’s” PSU?
Attachments
GB Pre-Order Status:
gary s / 2 / Australia * Pre-Order Placed *
mordikai / 2 / USA
Oracle1 / 2 / Greece * Pre-Order Placed *
Francois G/ 2 / USA * Pre-Order Placed *
Kokanee / 2 / Canada * Pre-Order Placed *
Zei / 2 / Sweden
astromo / 2 / Australia * Pre-Order Placed *
routhun / 2 / USA * Pre-Order Paid *
wtnh / 2 / USA
MEGA_amp / 2 / USA * Pre-Order Paid *
meanie / 2 / Singapore * Pre-Order Placed *
dBel84 / 2/ USA
................
xrk971 - thanks for your effort
gary s / 2 / Australia * Pre-Order Placed *
mordikai / 2 / USA
Oracle1 / 2 / Greece * Pre-Order Placed *
Francois G/ 2 / USA * Pre-Order Placed *
Kokanee / 2 / Canada * Pre-Order Placed *
Zei / 2 / Sweden
astromo / 2 / Australia * Pre-Order Placed *
routhun / 2 / USA * Pre-Order Paid *
wtnh / 2 / USA
MEGA_amp / 2 / USA * Pre-Order Paid *
meanie / 2 / Singapore * Pre-Order Placed *
dBel84 / 2/ USA
................
xrk971 - thanks for your effort