aleph xa critical performance line

[elviukai]

hi ,
lastly I powered my ultra high bias aleph x monoblocks

here is data of each amplifier-
tansistors 24pieces
bias 12A (very much I know)
rails at fets 19.2V
rails after rectifier diodes 22.4 under load.
each amp dissipate arround 580W..
transformers are rated 1700W each(manufacturer says so. has each secondary rom 3x 2mm wires)
I use CRC made from 12x 10W vishay dale resistors on each rail.

rectifier made from mached irf ulktrafast ultra soft recovery diodes . 600V 25A I made double bridges 8pieces per monoblock
mountened on pure cooper heatsinks bolted to chasis. in theory it shoul be able to handle 50A continiuos current.

I mached transitors very carefuly,also source resistors 0.5% tolerance and 20w. I get very close all readings between all pins of resistors from 0.482 to 0.488V.
absolute dc ofset is 0.2V between outputs 80mV (I will adjust it later)

my conserns-
theese mosters inspite only 80w/8ohms and 192W 3ohms power
gets very hot



aqt least hotest I have ever build

after 2hours I getting folowing temperatures

direcly on metal base of fets 95-105C'
heatsink at hotest place is 63-67C'
heatsink at coldest place 45-50C'


directly on diode metal base 100'C !!!! even 110C'
directly on diode heatsinks 90C'
CRC resitors has 140-150C'!!!!

the trafo also gets very hot arround 65C' its is heating itself not from other parts because it heats up when taken out of chasiis


also I get some strange feeling that something wrong with amp sound

playing at loud volumes piano makes some high frequency discorted sound(like tweeter resonates)
and belive or not this was fixed when I taked out L pad tweeter leaving just resistors in series (my xover is 1st order at 2000hz series filter )
I get back to my other aleph x monos (22V rails, 16fets , 8A bias) and they play fine in both L pad or just series resistors.

do my new amps do not handle my xover???


also forget- power suply at the moment is wery weak only 350 000uf. (will be 650 000uf) maybe this cause.

also I have complains about theese ultrafast ultrasof recovery diodes. - they sound very analytical bet very dry and harsh with my speakers(and its very noticeable)

simple diode bridge sounds more darker, smoother, liver.


where is problem???
also question is it ok to run amp devices (fets,diodes, resistors)
so hot?

[elviukai]

any one?

[Zen Mod]

Originally posted by elviukai
any one?

pictures ,actual schematic ....?
anyway-everything is way too hot ; don't even think about sound until you make it colder overall

[elviukai]

hi Zen mod, I did not put photo because its too big and ugly monoblocks if I will do there probably apear Choky with this "what! its uuuugly!!!"

now more seriuosly
the chassis itself is big,jeatsinks are big they are able to dissipate from 400 to 500W.

I did not stress mosfets too much -its 24 pieces irfp244 running 1A each gain mosfet. also one mosfet dissipate 19.4W -its not much.. but overal average heatsink temperature is 50C' (25'C above ambient)


what I am worry very much is diodes..

very bad that speakers are 3-4ohms should I reduce rails(to 17V) or bias(down to 10A)??

I looked datasheets looks like diodes and mostets can handle over 160C' but..

shematic is kristijan

just resistors are 0.5 and 24 devices also output to ground I changed from 33ohm to 50ohm(10W)

http://www.kk-pcb.com/alephx.html

[macka]

Hi,

Its difficult to know.

A few observations:

The transformer is getting too hot.

The diodes are way too hot

Try standard 25 amp bridges even in parrellel. You do not need massive capacitance as long as ripple current rating on the capacitor after diode is 3 times the current on reach rail. 40,000 should be plenty if you have CRCRC.

According to the spread sheet you should only require 9-10 amps of total bias.

Check all the wiring again.

It might be current limiting. Sounds like clipping. It thsi also at low levels?

Also make those 100 ohms current limiting resisters are the right value for your design. Blues hs good hand on that side of it.

Macka

[Zen Mod]

you know-this x is nothing more (and as usual with Papa-nothing less) than one plain and simple Alepx mirrored ;
so - I mean here primary on line of thinking-whatever your problem with this amp is,think in boundaries of "common" ALeph and I'm sure that in this way it will be easier to solve whatever problem is .

570 watts is bloody mucho of heat ,and I'm surprised you have doubts of this amp's current ability
anyway find enclosed (attached,whatever ) one nice pdf of our dear friend which name always slips from my mind ....he does nice work with this pdf ,same as recently with one pdf file for one Papa's phono stage

look at these values and think a little what else you can measure,statically and dynamically;
then write here what you measure,and we will figure out something,hehe-"with little help from our friends"

anyway-check current limit(er)s in both legs (left and right ,I mean ) and also check AC current gain

[elviukai]

thanks guys I will measure and write back..