Once rectified to DC, a 50vac transformer will be along the lines of 65 volts or so.
"Transformer Per Monoblock: Recommend 600VA or higher Transformer (example: Antek AN-6435) 35+35V secondaries"
You can go with an 800va transformer but they are specified to be oversized by Pass Labs. So if a 600va is what Pass Labs used, I wouldn't be worried about using something in that range. Capacitance is and a soft start circuit, something like that is probably a better investment.
I wouldn't go up in voltage unless you have the heatsinks for it. It seems like only a 30% jump in voltage but that isn't 30% in power. Much more and much more heat. I would personally stick with the 35v so you can bias it heavily as intended. Biasing the MOSFETs heavy is what gives the amp part of its sound.
"Transformer Per Monoblock: Recommend 600VA or higher Transformer (example: Antek AN-6435) 35+35V secondaries"
You can go with an 800va transformer but they are specified to be oversized by Pass Labs. So if a 600va is what Pass Labs used, I wouldn't be worried about using something in that range. Capacitance is and a soft start circuit, something like that is probably a better investment.
I wouldn't go up in voltage unless you have the heatsinks for it. It seems like only a 30% jump in voltage but that isn't 30% in power. Much more and much more heat. I would personally stick with the 35v so you can bias it heavily as intended. Biasing the MOSFETs heavy is what gives the amp part of its sound.
Speaking of doughnuts... I drove by Randy's Doughnuts in West LA last saturday. I was gonna take a picture for you but I was using the phone for Navi and didn't have the time -the red light was for once.. too short!
There were a couple of people in line that sure looked like 125KVA transformers with legs. I guess eating lots of donuts makes your transformer bigger?
I did a little looking in old threads last night. 35+35V is off the shelf at Antek. That's the "easy button". If you want to get bang on voltage - 37+37V looks like the exact match.
Perfect. Looks like they have a 600va 38v transformer. I suppose if you had the heatsinking for it, that could work as well.
The aleph J build I completed had 18v secondaries. After rectification I only got 5 watts DC more on each rail. I will be using two 5u cases. I am much less knowable then most here. Could I push the heat sinks of 5U Monos to 38 Wats per rail on each mono? What would be the limit.
I played with the bias on the J and actually preferred the written suggestion of 400mv over increasing the bias higher. I played around a while before realizing that.
Thanks for all the help. I want to start sourcing my parts and get the project up and running.
I played with the bias on the J and actually preferred the written suggestion of 400mv over increasing the bias higher. I played around a while before realizing that.
Thanks for all the help. I want to start sourcing my parts and get the project up and running.
As best I can tell I'm burning ~66W per heat sink in a 3U x 300mm chassis, and after a couple hours at moderate volume I can keep my hands on the sinks for many seconds. Those are 0.4C/W heat sinks. I think the sinks in a 4U Deluxe (also 300mm) sink 0.31C/W.
FWIW I'm a fan of the sound of IRFP240's run hot & hard
It's funny you mentioned the F6 as it keeps popping up on my radar. One was raffled off at BAF 2022 & I happened to be reading the build thread for the clone. It would be a dead simple build & swap for me but alas 
I have a Sony VFET amp in the works and I just built a 100V bench supply for one of Papa's SIT L'amps. An embarrassment of riches!
I have a Sony VFET amp in the works and I just built a 100V bench supply for one of Papa's SIT L'amps. An embarrassment of riches!
The f6 is great amp. The Aleph has a much sweeter sound which can be very addictive. The f6 is more of a neutral sound.
I had the f6 for a couple years when I built an Aleph. I had a hard time shutting the Aleph off even though it didn't have the bass control that the f6 had. Both are great. However, they are just great at different things.
If you're looking for another amp to build, Zen Mod praises the M2.
I had the f6 for a couple years when I built an Aleph. I had a hard time shutting the Aleph off even though it didn't have the bass control that the f6 had. Both are great. However, they are just great at different things.
If you're looking for another amp to build, Zen Mod praises the M2.
I just removed the resistors for a pair of devices so am running 4 MOSFETs now. It's just warming through. Until I get some more sinking this will be how I run it. I'm awaiting a quote for one of the wind tunnel type setups.
What is the failure mode of an Aleph and should we be running speaker protection?
My brother with mighty costly commercial kit is interested to listen to it in his system but I can't afford to replace his speakers if something badhappens!
What is the failure mode of an Aleph and should we be running speaker protection?
My brother with mighty costly commercial kit is interested to listen to it in his system but I can't afford to replace his speakers if something badhappens!
What do you feel are the benefits of running them hard?
For me the tone got sweeter but I lost the amazing sound staging width and depth especially. For my tastes the Aleph J is perfect running at spec. Didn't need any more in the tone area. I am interested to hear your thoughts on running the amp hot.
Thank you @Mikerodrig27 for inspiring the creation of a "Aleph Mini" build doc. And also for combing through old schematics / threads, and for input into and review of the document. Much appreciated!!!
The Aleph Mini build doc is now in post #1 of this thread. The amp is good for a 3U chassis. You can use donuts with 12+12 or 15+15V secondaries. The amp uses one pair of MOSFETs (1 output + 1 current source) per channel. Output is about 10-20W range depending on the rail voltage and bias. See the build guide for more info.
If you don't need big power, and you want the Classic Aleph sound - try a Mini!
The Aleph Mini build doc is now in post #1 of this thread. The amp is good for a 3U chassis. You can use donuts with 12+12 or 15+15V secondaries. The amp uses one pair of MOSFETs (1 output + 1 current source) per channel. Output is about 10-20W range depending on the rail voltage and bias. See the build guide for more info.
If you don't need big power, and you want the Classic Aleph sound - try a Mini!
Sorry I missed this earlier.
So, this is totally subjective, but to my ears, when an IRFP240 amp is run 'hot', the sound opens up. I can't really describe it better than that, other than it blooms, or seems full. All of the amps I've had with these devices as output stages, and even ones with them as supporting players like the F1J, just sound better after an hour or so of playing, which I associate with hot output devices. I built MoFo monoblocks with IRFP240s and inadequate heat sinking, and when the heat sinks were too hot to touch for even a second, that's when the amps sounded best to me. I didn't kill the devices running them that hot either, I re-used them in my Arch Nemeses and as far as I know they still work, so I have little fear in running them on the hot side. Quite the opposite to the LU1014Ds, I'm pretty sure sustained heat will kill them over time.
Any info I could look up to support this? When comparing data sheets, the junction temp ratings are the same between Vishay 240s and the LU1014D. However, I may have missed a key piece. I'm using those in a few amps, so any information would be very helpful. I know XRK pushed them 'reasonably hard', but that's all the info I've found thus far. Were they a known failure component in older amps etc?
Thanks!
I would be careful to mount them in a way that the heat they generate is fully sinked. Their drain plates are smol compared to the TO-247AC devices & hence the surface area available to dissipate the heat away is less. The XRK designed boards, I believe they are called IMS, seem to work just fine in my hands if you have soldered the drain plates properly- if you're using a hot plate or rework station to solder them, it can be hard to tell if the SMT solder under the drain plate has in fact fused. That said I use four LD1014Ds mounted to the IMS plates in my F1J clone at a Vds of 4.3V and I don't worry about them. It's when mounting them directly to PCBs or perhaps clamping them to heat sinks that I'd be cautious about heat transfer.
In the Zen Variations 9, Papa sez "The Lovoltech LU1014D requires significant heat sinking also, as will operate at 4 to 6 watts continuously." I think in one of these threads somewhere he cautioned against dissipating more than 5W through these things.
^ Thanks so much! I'll keep digging around. I was aware of the need for proper sinking, and I hope I've taken care of that. What I was looking for more information around was your comment that you thought "sustained heat will kill them over time" specifically when compared to something like a 240. I know dissipating the heat due to the small(er) package might be a concern, but I'm more interested in operating temperatures of concern over time. Like, how high is too high over time? As an example, I have no concern with running 240s at what I'd assume generally to be 75C+ junction temps for the life of my amps, but if I should not run the 1014Ds at those types of temps, I'd very much like to know.
I'd be grateful for any other information from all sources. Sorry for a bit OT.
Edited to correct - package to junction... key difference. My package temps are generally no higher than 60C.
I'd be grateful for any other information from all sources. Sorry for a bit OT.
Edited to correct - package to junction... key difference. My package temps are generally no higher than 60C.