L20 V8

[Ian Finch]

Maybe those Italian designers should tell us about this Ebay one here now. It will be interesting to see how stable these 4 pairs CFP are with pro. use (if they survive)

I have read (but not built) large CFP stages that are more stable because Re is kept in Emitter leads, to avoid inductance in collector leads and those connections can be short tracks. Otherwise. in regulators, you often see ferrite beads to damp oscillation there.
So you know the designers are aware of the problem with collector inductance.
The crazy thing is, most manufacturers try to solve it the same way as EF type and I don't believe it works that way. I can be wrong, too

[sakis]

ημμμ that can also be intersting i could look to a schematic if you have one

[ljm_ljm]

Originally Posted by sakis

ημμμ that can also be intersting i could look to a schematic if you have one

It is a standard, ordinary CFP structure.

Reverse darlington. The IRF610 IRF9610 driver.

There is also a standard EF output stage. It is suitable for most DIYER.
I will be his name out of L20V9。

[daviddsailor]

I powered up my L20 V9 boards today. Whilst testing at low power into 6 ohm resistors (about 10WRMS) I switched from sinewave (at 1kHz) to square wave. I blew both 2A protective fuse to the board. I found that one 817 and one 1047 were shorted. I took them out and used only the remaining 3 pairs. It works OK. I ordered one of each from a local supplier, cost $20AUD. Now I find I could have purchased 5 pairs for the same cost from the US!! Still I couldn't wait.
Now I have both L20 boards working I can see crossover distortion on the CRO waveform (tested to 35W so far). I would like to increase the idle current to at least 100mA. At present it is 10mA or less. Any suggestions? I probably can work it out myself but would like to know if anyone else has tried. I haven't listened to the sound yet. A project for tomorrow! Would be nice if I had a circuit. The V9 uses a mosfet driver. The circuit I have is for an earlier version.

[rickedd46]

The seller should be able to provide the schematic for you. Did you ask?

[daviddsailor]

The supplier of the amp boards gave me a circuit which turned out to be an earlier version. I tried again and received a circuit which has the mosfet driver but only shows the front end as labelled boxes. I have assumed that if I cut and paste the circuits I might end up with the correct one for the V9, but somehow I don't think I will ever trust it unless I trace everything out. The designer of the L20 hasn't responded to requests for the circuit. There seem to be more V's of the L20 than I've had hot breakfasts and I find the whole issue quite frustrating (as one who likes things simple!).

[DIYGEEK]

I've been looking at the L20 for some time now the number of transistors and output wattage is impressed me. But I'm looking for something really stable and has protection. I am happy you posted this discussion of the L20, think I will hold off on purchasing one. Pat

[daviddsailor]

My Ebay power supply boards arrived recently. Fitted with suspicious looking Nippon Chemicon 80V 10,000uf electros type KMG code A7088. I can't find any NC 10,000uf KMG types in NC current datasheets. Recent production KMG types are radial lead in much smaller values, not snap in types. Am I being paranoid about these caps? Do they fit the "bad caps" mold? Comments ...

[carlmart]

Originally Posted by ljm_ljm

It is a standard, ordinary CFP structure.

Reverse darlington. The IRF610 IRF9610 driver.

There is also a standard EF output stage. It is suitable for most DIYER.
I will be his name out of L20V9。

I wonder why the schematic for this amp is not being disclosured. Particularly if there seems to be several versions of it.

Diygeek has already held off on his potential purchase, and there will be many others that will do so.

I am involved on another thread over the Diamond Differential kit sold by Jim's Audio and Tubeshunter, and from what I see the only reliable eBay kits seem to be those involving 3886 chips.

OTOS I think discrete amps can be tweaked better, and this discussions should help people that are looking for discrete affordable kits on eBay that DO WORK AS THEY SHOULD!

Why don't we start here in doing so?


Carlos

[daviddsailor]

I got back to the L20 boards again yesterday. I replaced what was left of the output devices on my blown up board with 4 pairs of new devices. Everything worked OK initially except that they have a rising frequency response with twice the voltage out at 100kHz than at 1kHz (my generator doesn't go any higher). I tested both boards to 40WRMS into 6 ohms then I noticed both amps getting very hot without any input. One in fact had blown the 2A protection fuses. So I guess they went into HF oscillation even though I didn't see it on the CRO. Unless I'm doing something radically wrong, these boards are hotter that a two bob watch. Today I will put an inductive LPF on the outputs and see what blows up next!