X-Amp (LME49830 + Exicon Lateral Mosfets)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I've assembled two amplifiers now but I haven't powered them yet. First I need to figure out how much quiescent current I should bias the MOSFET's with and appropriate resistor values (R7 and R8). I'm thinking 40-50 mA per pair @ +- 60V.

Here are some pictures of the current progress.
 

Attachments

  • DSC_0360_.JPG
    DSC_0360_.JPG
    235.7 KB · Views: 1,217
  • DSC_0371_.JPG
    DSC_0371_.JPG
    130.6 KB · Views: 1,898
  • DSC_0373_.JPG
    DSC_0373_.JPG
    189.1 KB · Views: 1,119
  • DSC_0377_.JPG
    DSC_0377_.JPG
    186.5 KB · Views: 1,171
So I soldered on a 2k pot where R7 is supposed to go to find an appropriate value(s) for the biasing resistor(s). I preadjusted the pot to 300R which I figured would be safe (0.002*300=0.6V) and just to be extra safe I set the lab supplies current limit to 150mA/rail. I started turning the pot slowly and nothing happens and so I turn some more and nothing, I then measure the voltage across the pot to find it to be 2V and still nothing, no increase in current consumption and no voltage across the source resistors.

Hmm.. the voltage between POut and NOut is 2V and so the MOSFET's should start to turn on right.. after some headscratching I measure the voltages on the N-channel gates relative to ground to find them to be -1V and of course the voltages on the P-channel gates are +1V... curses!

So I managed to mix up pin 13 and 14 on the LME chip schematic symbol, I guess I was thinking POut - positive rail and NOut - negative rail not POut - P-channel and NOut - N-channel as I should have been...

Desoldering pin 13 and 14 and switching them around was pretty easy though. Luckily the mixed up pins are next to each other it would have been uglier if they were on opposite sides of the chip.

With the pins soldered in the correct places I got 45mA per pair with the pot set to 510R so no need to use R8 (the purpose of R8 is to be able to fine tune the quiescent current by parallelling two resistors to get a wider range of possible values).

Anyways the amp has been playing for a couple of hours now and I'm quite impressed it sounds pretty good.
 
Source for those lateral FETS in USA

You can buy the identical lateral fets form Newark and they are in stock.
Part number is ALF16N20W and ALF16P20W. They are from Semelab. Datasheet is identical to the ECW20N20, ECW20P20. You can get TO-3 packages if you are into metal.....

BTW, there is NO gate protection on these devices. Read the data sheet. I'd recommend using 12V zeners for G-S over voltage protection. But I see you have some 8.2V zeners.....

Anybody have Spice models??? I have some that "work" in LTSpice but since I created them I'd rather have the real things.

The 20A (or 16A as called by Semelab) devices I believe have 2 dies in parallel inside each TO-264, based on the power dissipation doubling as the device goes from 10A to 20A. Also, the input (G-S) capacitance is almost double of the 10A device.
 
Last edited:
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.