I have had a JLH running pretty hot for a few days.
The tips of the heatsink fins are above the "burning fingers within 5seconds" test temperature.
I would guess the output devices, 1pr mj15003 are running at Tc~80°C and Ts~70°C
So far I have not detected any deterioration in performance.
The tips of the heatsink fins are above the "burning fingers within 5seconds" test temperature.
I would guess the output devices, 1pr mj15003 are running at Tc~80°C and Ts~70°C
So far I have not detected any deterioration in performance.
I've got a couple of slow running 50mm fans which "waft" air throughout the body of the amp to keep the caps cool.
Inevitably they will pick up some heat from the MOS-FETs to keep the cat warm.
Inevitably they will pick up some heat from the MOS-FETs to keep the cat warm.
I opted to try to keep things a bit cooler for reliability reasons.
Have you tried some 12V fans running at say 5V. They are almost silent at that speed and make a huge difference to the efficiency of the heatsinks.
This is still in the "experimental" stage.
I can easily lower rail voltages and/or bias to suit the sink.
Experimental, because the output offset wanders massively before settling down to a nice low value. I will probably change to an AC coupled version on single polarity supply to completely avoid the offset issue. The output cap can be both a DC block and bandwidth filter for mid or treble drivers.
Sorry, this is substantially off topic but some of the content can be extended to F5 operation/s.
I can easily lower rail voltages and/or bias to suit the sink.
Experimental, because the output offset wanders massively before settling down to a nice low value. I will probably change to an AC coupled version on single polarity supply to completely avoid the offset issue. The output cap can be both a DC block and bandwidth filter for mid or treble drivers.
Sorry, this is substantially off topic but some of the content can be extended to F5 operation/s.
I listened to a stereo version of F5T V3 lastnight. There is something we are all missing about it that was in the original article. The presence of the gives more Class A bias for a given setting than a standard F5. I am not exactly sure of the theory, but it is plainly stated by Nelson. Last night a listened to Magnapans with just a single output pair, and I find myself surprised at the ability of this amp to drive anything.
Myself and Buzzforb have a few iterations we are playing with, but I have not tried as a goal to push the heat. More to make sure the circuit is functional. It is, and that was my main goal.
There really have been no surprises, other than that the cascode set up takes some thought around biasing. So far we believe that higher IDSS JFETs in the V3 design makes it easier to get some range of bias out of the pots.
It sounds like you Aleph platform can handle the heat here, if not - just bias a bit lower.
Ill add another pair and let you know. I have to spend some time with it. It is very much an F5, but on steroids. Not that the standard was wimpy, but with 4 pairs or output devices, you gave a brute of an amp that seems to provide the same positive attributes of it's little brother. If you want fancy adjectives, ill have to pull out the dictionary🙂
Tea,
Sorry I haven't been in touch. Your boards have performed perfectly. I have found no significant different e in mounting the fets with the diodes, although I will probably operate them onto another heatsink on final version. Went ahead and signed up for final board revision. Thanks for the opportunity.
Tea,
Sorry I haven't been in touch. Your boards have performed perfectly. I have found no significant different e in mounting the fets with the diodes, although I will probably operate them onto another heatsink on final version. Went ahead and signed up for final board revision. Thanks for the opportunity.
Last edited:
I have beeen so impressed with the Aleph 4 that I'm wary about destroying it to convert it to an F5(T)v3.
My speakers are not a good load for an F5 so they might just work with the (Turbo) They certainly work well with the Aleph 4.
My speakers are not a good load for an F5 so they might just work with the (Turbo) They certainly work well with the Aleph 4.
What does this mean?
Have you changed the bias current? Or is it the same?
Have you set the Vrs high enough that the diodes are passing, while setting bias?
Have you considered that setting Vrs that high does not allow one to actually determine the bias current?
Or, did you mean that peak transient current delivery is improved because the diodes bypass the Rs?
Last edited:
then build a F5turbo from scratch 🙂
the best way if you want v3 is to build monoblocks.
i guess you can get heatsinks from bermingham aluminium for a good price.
I was trying to save a few hundred pounds.
OK in a few months time it might be better to keep the Aleph4 intact and start again.
OK in a few months time it might be better to keep the Aleph4 intact and start again.
Don't worry, its messing with me too. We are missing something about how this functions. I had a conversation with Jacco about it, but can't find it. I think.it is in the original F5T posting thread.
K&D,
I can assure you that the V3 version will have no problem win your B&W's. I would expect it to be more capable than Aleph 4. Whether you like it more or not, dunno.
If the speakers are likely to demand high peak currents then balanced and bridging is not the solution.
The paralleled output devices is the way to meet the current demand of the hard to drive speakers.
The paralleled output devices is the way to meet the current demand of the hard to drive speakers.
I think that many are not recognising what the diodes bring to the amplifier operation.
It was thinking about this that led to my reservations on how to set up and operate the diode versions of the amplifiers.
I now think that the lack of responses to my call for information is because the majority still do not realise what the diodes do !
It was thinking about this that led to my reservations on how to set up and operate the diode versions of the amplifiers.
I now think that the lack of responses to my call for information is because the majority still do not realise what the diodes do !
That is precisely why the Aleph 4 won it when I put the Alephs to the test against my B&Ws.