But if I were to take a wild guess as a starting place, lhquam's graph makes me think something in the neighborhood of 1.3 amps and between 1.3 and 2 volts might work.
I am new to all this SIT business. So is the LU1014 kind of like a SIT in behavior based on these graphs?
here, LU is cascoded, practically being different (compound) animal
nice to think of it as triodish, but ....... nope
btw. same would happen with SIT, if cascoded
nice to think of it as triodish, but ....... nope
btw. same would happen with SIT, if cascoded
Here is a simulation of a 100 Watt (into 8R load) amplifier using a single LU1014D dissipating only about 6.6 Watts at idle. The results are hard to believe.
Attached is the fourier listing:
Attached is the fourier listing:
Attachments
Looks beautiful!
Worst case?: "0.01° Total Harmonic Distortion: 0.032241%(0.032338%)"
That's remarkable.
And just poking along at ~3 Amps...doesn't look like it'll be stressed.
Worst case?: "0.01° Total Harmonic Distortion: 0.032241%(0.032338%)"
That's remarkable.
And just poking along at ~3 Amps...doesn't look like it'll be stressed.
Looks like super result. But why are we paralleling 4x TO-247’s when the little pinner LU takes the full brunt of all that current?
I have used these before. They are small and do not have sufficient surface area to extract heat effectively.
I have routinely pushed 150w from a single IXYS TO-264 MOSFET with a heat pipe CPU cooler and Noctua fan. That’s just one device. The CPU coolers are very effective.
This was a 50w MOFO:
Last edited:
I would go with lower Iq, allowing for some KlunK!
dynamic current peaks still same, but no need for that amount of heat
I mean, not that there is no need but beat me ........ I'll go SUSY to keep my but less sweaty
remember F4
just 160W of heat for 100W monoblock
dynamic current peaks still same, but no need for that amount of heat
I mean, not that there is no need but beat me ........ I'll go SUSY to keep my but less sweaty
remember F4
just 160W of heat for 100W monoblock
Xrk971:
You ask some good questions:
You ask some good questions:
- I suspect there might be FETs capable of 150W+ idle power dissipation, but extraordinary measues are required to carry the heat away from single devices. BTW: what IXYS FETs did you use?
- By paralleling 4X TO-247 FETs I limit the idle power dissipation to under 45W each and spread the heat across a "more normal" heatsink.
- You are correct to be concerned about all of the current through the tiny leads of the LU1014D.
- It should be deal with 6.6W-10W idle power dissipation of the LU1014D.
- Nelson used goop to anodized aluminum for the F3.
- F3 cloners soldered them to copper spreaders.
- Another option is goop to a copper spreader atteched to the heatsink with good and insulator.
well, soldering little critter to copper plate should be easy, using Hot Air station and , of course, using tricks as preheating copper before positioning LU on
I'll stop here, at least until I finish with ongoing amps projects ..... only then I could think about Ultrasymmetry LU amp
🙂
bummer, I don't think I need LU for that one, at all
I'll stop here, at least until I finish with ongoing amps projects ..... only then I could think about Ultrasymmetry LU amp
🙂
bummer, I don't think I need LU for that one, at all
Zen Mod, are you still planning to sell kits for the LuDEF? I definitely want one but am in NO rush (currently working on SissySIT). It should be a very cool project. If you have the circuit finished, I can take a shot at designing the PCB layouts - it won't be as good as yours but might work...
now, after several posts I wrote, being looped with bloody LU , finally stepped aside and realized one thing must be asked :
in what, except irresistible quirkiness, is that better than IXYS puck follower?
be sure - when you see that I posted all pics as I did for both LuDEF and Old Soul, that project ( hate that word, it's so modern nowadays, labeling so many specimens of pure trash) ...... anyway - that project is finished
Yes, I'll have kits soon ; have enough LUs to send with kits for at least 35 stereo packages
Catch is** - now I'm in the loop of working on finishing ( damn, yes, all tried and confirmed in prototypes and all having finished pcb files) some 15 different amplifiers, more or less different and more or less similar, but in most cases being worthy of realization
2 of them already shown - LuDEF and Old Soul; SissySIT R.3 in few days, as soon I put already populated pcbs in nearest cradle having some sinks and some PSU
** so Mighty ZM being confused,to live in outer world and work on logistics for kits, or to proceed with further amps
I haven't yet seen a way to scale up the XA25 to higher power levels without degeneration.
BTW: I want to buy an LuDEF kit too.
I ran the same simulation as before at 200W:
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 1.000e+02 5.594e+01 1.000e+00 -0.00° 0.00°
2 2.000e+02 8.962e-04 1.602e-05 -102.29° -102.28°
3 3.000e+02 5.292e-02 9.460e-04 0.46° 0.47°
4 4.000e+02 1.684e-03 3.011e-05 82.37° 82.37°
Total Harmonic Distortion: 0.094663%(0.095558%)
Harmonic Frequency Fourier Normalized Phase Normalized
Number [Hz] Component Component [degree] Phase [deg]
1 1.000e+02 5.594e+01 1.000e+00 -0.00° 0.00°
2 2.000e+02 8.962e-04 1.602e-05 -102.29° -102.28°
3 3.000e+02 5.292e-02 9.460e-04 0.46° 0.47°
4 4.000e+02 1.684e-03 3.011e-05 82.37° 82.37°
Total Harmonic Distortion: 0.094663%(0.095558%)
Attachments
we'll work on scaling up, in few months
purely as mental exercise, being aware that even 15W is more than I ever would need
purely as mental exercise, being aware that even 15W is more than I ever would need
I agree that it is a theoretical exercise.
The situation with the my SIT3X was different. Those 2SK182ES SITs were crying for higher power levels than 25W.
The situation with the my SIT3X was different. Those 2SK182ES SITs were crying for higher power levels than 25W.
I still wonder what Pa was thinking when running these tests with cascoded IXYS FETs. The Aleph Design Reloaded
