N-channel amp.

Status
Not open for further replies.
disaster area

Picture of the remains...

Arne K
 

Attachments

  • n-ch-fire2.jpg
    n-ch-fire2.jpg
    79.7 KB · Views: 1,347
Re: More q...

Cobra2 said:
Can anyone guestimate the output power w. +/- 60V?
Arne K
[(0.7x60v)^2]/8 ohm= 225 watt

if you want 100watt/8 ohm
SQR(100x8)x1.4= 40 volt peak = +-40V
80 watt /4ohm
SQR(80x4)x1.4= 25 volt peak

Of course PSU can sag some 5 volts, so you have to add for that,
and some 2 volts for Transistors workspace
 
Cobra2

You have mounted the IRF610 Thermal sencing device incorrectly to the o/p stage. Mounting it on top of the PCB like you have done is very bad and you can see the result.
It must be in direct contact with the top of the o/p device and I see you haven't added the Source resistors or the rest of the manatory upgrades, that this amplifier requires for thermal stability.:bawling:

Please read the manual again, its all in the construction manual.

Regards

Anthony Holton
www.aussieamplifiers.com
 
anthony my amp is looking difrent now the reisters that you told me to put on are on now that pic that i posted was just a pic just because ihad a pic of what i had done so far a whyle back. also humm if these transisters have to be mounted ontop of the fet then why did you design it this way..?

cant i mount the fet to the heat sink. the irf610 the bias one. ? the one that controles the bias from heat.

?
 
Jason That photo of my amp is an old one as well. However it does have the thermal sencing device mounted on top of one of the fets.

Mounting on the heat sink will work OK, but don't increase the rail voltage. The Gate theshold voltage at higher supply voltages just gets to much out of control, without further modification of the thermal tracking senser.

Anthony
 
well actualy longer than that ebcause there going to anthony not me : O ) but ill get my next parts shipment from him with the boards eeh

just ordered heatsinks for my aleph 2's yeah.. now to start saving for toroidals
i need 4 x 1 kva 2 for av800's and 2 for alephs there is $500.00 right there.: O ) man this bhobbie is expencive.

J'
 
here we go...

How do we adjust bias without resistors (ref. the assembly-instructions).

Anyway, I must have misunderstood an earlier post where +/- 95V DC were mentioned, with some other mosfets, - (so I thought that it would be them living dangerous).

Glad I have spare PCB's, finished, all resistors matched to ~0,2%...
I'll just have to redesign PSU, and make better thermal coupling before I try again.

Arne K
 

Attachments

  • n-ch-resist.jpg
    n-ch-resist.jpg
    48.9 KB · Views: 1,378
Working!

N-channel, the sequel...

The previous experiment also killed all(!) the 0,33 Ohm (3 x 1 ohm-2W) resistors as well.

So some adjustments had to be done:
2 x New trafos, 2 x 38V gives me +/-55V DC,...updated "modified" PCB, (hacked out for thermal coupling of Q12).
Now everything seems stable.

Arne K😀
 

Attachments

  • n-chnl-xx.jpg
    n-chnl-xx.jpg
    97.2 KB · Views: 1,266
I can do THD analysis...

-If you will throw me some high-power load resistors...

Sound-vice, you do not need to worry about distortion.
It sounds very clean (and silent), with a strong and powerful bass. Very well controlled.

I think I will give it a 96% score against my14 year old Dynamic Precision Amp, ( 2 x 250W Class A), maybe a 115% score in bass area, (probably caused by dried up caps in my old amp).

(BTW, I use 2 x 500VA trafos (dual mono) and 44 000uF pr.rail, total 176 000uF)

Arne K
NORWAY

(Can hardly wait to start building Aleph 2, for bi-amping...)
😉
 
Status
Not open for further replies.