I'm confused 
If I'm after a 32V rail voltage, what transformer secondary voltage rating do I need? I have read where there's a voltage drop from the transformer onward so can someone please explain how Nelson would get 32V rails from 24V secondary's as per the below quote please?
QUOTE from paper on the F5 & F5T by Nelson Pass..
Increasing the power supply voltage is the obvious way to get more power out of an F5. You can simply raise the supply rails to +/-32 Volts and get 50 watts into 8 ohms right away without other modification. 24 V AC secondaries on the power transformer will do it. Don't forget to use higher voltage power supply capacitors.

If I'm after a 32V rail voltage, what transformer secondary voltage rating do I need? I have read where there's a voltage drop from the transformer onward so can someone please explain how Nelson would get 32V rails from 24V secondary's as per the below quote please?
QUOTE from paper on the F5 & F5T by Nelson Pass..
Increasing the power supply voltage is the obvious way to get more power out of an F5. You can simply raise the supply rails to +/-32 Volts and get 50 watts into 8 ohms right away without other modification. 24 V AC secondaries on the power transformer will do it. Don't forget to use higher voltage power supply capacitors.
The 24V secondary is 24VAC, which when rectified and filtered gives around (24 x sqrt2) or 33V DC.
It helps clarify to include the AC or DC when stating voltages.
It helps clarify to include the AC or DC when stating voltages.
Last edited:
thanks kindly... will state ~ or +/- next time.
My plan is to first build an F5 then eventually up the rail voltage to get the 50W out, so knowing what transformer I need was um.. important.
Thanks again
My plan is to first build an F5 then eventually up the rail voltage to get the 50W out, so knowing what transformer I need was um.. important.
Thanks again

If you use the same heat sinks, then to keep the same sink temperature you would have to
decrease the idle current by a factor of (24/32), which is 2/3 that of the standard circuit.
decrease the idle current by a factor of (24/32), which is 2/3 that of the standard circuit.
Last edited:
Maybe, but I'd check on this forum for what people have already dome successfully. Maybe they just
lower the bias current and use the same heat sinks. There are larger sinks, though.
The F5T uses multiple devices per phase per channel, and spreads out the thermal load on the sink
better than the two devices per channel of the standard F5.
lower the bias current and use the same heat sinks. There are larger sinks, though.
The F5T uses multiple devices per phase per channel, and spreads out the thermal load on the sink
better than the two devices per channel of the standard F5.
Last edited:
OK thanks for raising this Rayma, I think the best way forward given I want to go the upgrade path is to make the F5T chassis right from the get go. Have an ol' buddy who just so happened to inherit a machine shop, its chockers full of milling machines & CAD cutting gear.
So are there any negatives if I over heat sink the chassis and not run the trannies to ~20 degC / ambient as maybe the case with the initial F5 setup?
So are there any negatives if I over heat sink the chassis and not run the trannies to ~20 degC / ambient as maybe the case with the initial F5 setup?
Last edited:
Now that sounds like a plan. Do it right the first time.
Have him mill the sink mounting surface nice and flat like a mirror, for better heat transfer.
Then use spring mounting clamps for the devices, rather than bolts.
Running electronic parts cooler won't ever be a negative, other than cost and size.
Thought about monoblocks? For your situation the extra cost would only be another transformer
and PS board. A no brainer.
Have him mill the sink mounting surface nice and flat like a mirror, for better heat transfer.
Then use spring mounting clamps for the devices, rather than bolts.
Running electronic parts cooler won't ever be a negative, other than cost and size.
Thought about monoblocks? For your situation the extra cost would only be another transformer
and PS board. A no brainer.
Last edited:
Yep, agree on flat surface for better heat transfer, will be going bolts though to keep things simple.
Yes I have thought of monoblocks, have always wanted a pair which is why I'm going the ACA route first up.
I plan to first build a stereo ACA then grab another ACA & go monoblock, that way I can see first hand what monoblocks bring to the table for me... beyond balanced to single ended.
So I guess watch my space.
Yes I have thought of monoblocks, have always wanted a pair which is why I'm going the ACA route first up.
I plan to first build a stereo ACA then grab another ACA & go monoblock, that way I can see first hand what monoblocks bring to the table for me... beyond balanced to single ended.
So I guess watch my space.
Last edited:
Rayma, so in comparison beyond the balanced thing whats the advantages of monoblocks please?
Really I think for my first few DIY amps its best I keep things relatively simple.
I do have an electronics background in gyro systems ect that hasn't seen the light of day for 45 years, rebooting an ol' brain aint that easy
Really I think for my first few DIY amps its best I keep things relatively simple.
I do have an electronics background in gyro systems ect that hasn't seen the light of day for 45 years, rebooting an ol' brain aint that easy
Last edited:
Monoblocks improve resolution and clarity, compared to a stereo amp with a shared supply.
However, a bridged "balanced" amp has two out-of-phase channels connected in series and
goes backward in those respects, even though it can have more power.
Monoblock (but not balanced) amps are simpler. Just build one, and then repeat. Lighter on
the ol' back also. And they lack the inherent ground loop that a stereo amp with a shared supply has.
However, a bridged "balanced" amp has two out-of-phase channels connected in series and
goes backward in those respects, even though it can have more power.
Monoblock (but not balanced) amps are simpler. Just build one, and then repeat. Lighter on
the ol' back also. And they lack the inherent ground loop that a stereo amp with a shared supply has.
Last edited:
makes good sense to go monoblock off the bat.
I guess you'd run the former secondary's in parallel?
What transformer VA would suffice for a F5 25W ~33VDC rail monoblock please?
I guess you'd run the former secondary's in parallel?
What transformer VA would suffice for a F5 25W ~33VDC rail monoblock please?
Last edited:
I would consider 400VA to start.
Let's say you're building something like the F5 turbo V1 on page 4:
https://firstwatt.com/pdf/art_f5_turbo.pdf
and you have enough heatsinking and you bias it to 50W class A into 8 ohms.
That needs about 1.8A per rail so about 120W total dissipation for the
channel.
A 400VA should be nice and cool at that load and still gives some room
for higher bias. (You probably shouldn't go much more than 40W per
device or ~160W dissipation per channel)
And you can of course go higher. You'll have to consider the balance
between availability/cost/size/weight. 🙂
Let's say you're building something like the F5 turbo V1 on page 4:
https://firstwatt.com/pdf/art_f5_turbo.pdf
and you have enough heatsinking and you bias it to 50W class A into 8 ohms.
That needs about 1.8A per rail so about 120W total dissipation for the
channel.
A 400VA should be nice and cool at that load and still gives some room
for higher bias. (You probably shouldn't go much more than 40W per
device or ~160W dissipation per channel)
And you can of course go higher. You'll have to consider the balance
between availability/cost/size/weight. 🙂
- Home
- Amplifiers
- Pass Labs
- Transformer secondary's required for grid voltage