LuDEF

Jump to Latest
#221
Zen Mod said:
worth looking in Papapatents, as first place

benefits you're expecting?
Click to expand...

One could build a 100W/ch, 200W/ch, or more amplifier without the degeneration using the single-cascode multiple-FET topology of post #200. Not that I need such a monster, just a theoretical exercise. I have no idea how far the LU1014D could be pushed. Perhaps to 3A idle current, which would be enough for 100W.
 
#223
not that I would trust in those tiny pins, for anything substantial dynamically

100W means 80Vpp envelope, which means peak current in range of 5A with benign 8R load

call me a Chicken, but without at least 2 pcs of LU in parallel, I wouldn't even try 100W amp, more for more

of course , one can make non-degenerated Stasis OS, using tiny buggers as drivers/fillers
 
#225
Running that much current through these things will require better heatsinking. They are tiny - much smaller than TO220. JPS64 and I are thinking of making an aluminum core PCB adapter that will allow these LU1014’s to be mounted and resemble a TO-247 package. It will provide a heat spreader to broaden the area for thermal flux to pass to the heatsink.

5A x 20v drop is 100w dissipation. I wouldn’t want to put more than 20w each on these so we are talking 5 LU1014’s in parallel for a 100w amp.
 
#226
@lhquam

well, that's exactly the trick ..... I mean - no trick but catch

of course, I didn't finished the sentence - really don't know how to parallel them without Rsource and is it even possible, even with all precautions in matching ...... thing which needs to be tried in real life

when I mentioned Stasis,this came to mind - adequately biased NP pair (of JFet + Mosfet) but also just one pair of big mosfet outputs, which is possible if we use big Pucks

at least logical for me - when going Square Law route, paralleling is either tricky or impossible, so using single pairs is pretty much sole viable solution

in this moment I see it as working, dunno later :rofl:

frankly , not soaked enough with coffee today......
 

Attachments

  • Bstrd Stasis.png
    Bstrd Stasis.png
    8.9 KB · Views: 420
#227
xrk971 said:
Running that much current through these things will require better heatsinking. They are tiny - much smaller than TO220. JPS64 and I are thinking of making an aluminum core PCB adapter that will allow these LU1014’s to be mounted and resemble a TO-247 package. It will provide a heat spreader to broaden the area for thermal flux to pass to the heatsink.

5A x 20v drop is 100w dissipation. I wouldn’t want to put more than 20w each on these so we are talking 5 LU1014’s in parallel for a 100w amp.
Click to expand...

I'm not so scared of heat, as I'm scared of current through tiny bugger

for all sane currents, heatsinking is adequate and all you can do with any other arrangement is just going to ease mounting

not that that goal isn't worth pursuing
 
#232
Thinking about current hogging and bond wires, couldn't a parallel/tripled/etc F3 be done with separate output caps for each F3 clone, and tying them together after the caps?

I'm definitely interested in a group buy for the F3, by the way 🙂 Our F3 clone is wonderful, but I've been playing with low resistance drivers lately

(On a similar note, maybe a F8 turbo could be done the same way)
 
Last edited:
#233
well, make your mind - either low/semi power A class amp, finely tuned as Minarelli 50cc ...... or big honking/low rails/high current amp

hard to have everything ....... big pucks are not liking low voltages, IRFP sorta same , LU right there, but wimpy wires .......

of course, if you introduce output xformer, The World is your Oyster

it seems changing back to sane impedance drivers is simplest

:clown:
 
#234
Point well taken, I'm making a 4 channel speaker with the 6-24, so LuDEF for the woofers will have the clean power anyway.

...But if I could parallel the HF amps to play with loadlines or use them on the woofers..... (49% justification, 51% Greedy Boy 🙂

Edit:
I use some toroidal autoformer on the output to play around as well, but we have some ESLs that benefit from the current for HF capacitance. But those ESL's are justifying a HV Vfet amp plan already.
 
Last edited:
#237
lhquam said:
I just found this article about the current limits of MOSFET bond wires:
http://www.aosmd.com/res/application_notes/mosfets/Continuous_drain.pdf

Based on the LU1014D datasheet, the circular cross section of the leads appear to be about 0.4 mm^2 which corresponds to about AWG 21 wires.
Click to expand...

American Wire Gauge Chart and AWG Electrical Current Load Limits table with ampacities, wire sizes, skin depth frequencies and wire breaking strength says 21 AWG should be able to handle 9 Amps; another chart I got elsewhere says about 7.25 Amps.
 
You must log in or register to reply here.
Top Bottom