JAT EZ amp - idea by John Audio Tech

[chermann]

played around...

good night
chris

 

[chermann]

so i have 4 open issues/questions:
1. CCS with LED can i use a smal cap to stabilize
2. for R6 (LTP) can i use 2 parallel resistors with about 1k5 to have a better temp stability? or use a 1k5 metal film and a 1k5 carbon film
3. T5 according to my sim i got about 37,7V swing and about 7,5mA current that give 282mW. keep the idea to use a BD139 or use a "better" To92 , e.g. BC639(800mW) with heatsink - keep the 2N5551 with heat sink
4. recalculate the values as minek wrote for VAS , R11, R10, RV1 -in real live RV is a pot with 200R.
the Re of T5 must be recalculated

anything else?

kr
chris

 

[chermann]

point 4:
i just rise a bit the Re of T5 to 13R ---> BDW94C (PNP) 73,7mA - BDW93C (NPN) 74mA
I R13 8,2mA
Ie Q4 6,2mA
Ie Q5 8,3mA

 

[chermann]

to point 1:
LED green /LED red
i found this comment by ostripper at the JAT 501 thread: here
Green LED works good , red led + 330R (R4) is more thermally stable.
so for the EZ amp its shold be okay

open is :
point 2
point 3

 

[chermann]

Hi

2. for R6 (LTP) can i use 2 parallel resistors with about 1k5 to have a better temp stability? or use a 1k5 metal film and a 1k5 carbon film
carbon has a bad TCR and i will not use this...stupid idea
R6 is valued with 759R, at mouser i just found a 750R 1% 50ppm...10pcs 3,38Euro -MRS16000C7500FC100, digikey has the 759R at 1/4W rating but they are too expensive and not in stock, ERC55759R00DHEK600 or RNC60H7590DSB14...useless for this amp.
i will use what i have at home.

3. T5 according to my sim i got about 37,7V swing and about 7,5mA current that give 282mW. keep the idea to use a BD139 or use a "better" To92 , e.g. BC639(800mW) with heatsink - keep the 2N5551 with heat sink
i will try this heatsinks to cool down the TO92 as written in post 18:
TO-92 heat sink

or

aliexpress TO92 heat sink

next day i hopfully find time to build the first prototype

kr
chris

 

[stocktrader200]

To220 is around 70w dissipation. I used irf530 and Irf9530 with 35v rails into 4 ohms in the car for 15 years. (125W RMS into 4 ohms)
The amp is still working.

 

[chermann]

Hi

first PIO-Putting into operation.
i use a 0,65K/W heatsink for first try.

IMPORTATNT -signal GND and PWR GND must be connected with extra wire.
VAS is use Tohsiba TTC because it is isolated...may version at home for BD137/139 have the metal surface - and i am to lazy to use additionally a cutted isolator

DC offset is -204mV
red LED gives 1,72v so its too much -CCS gives 2,85mA

trimm pot with 200R is fine to adjust the bias to 50mA each no heat at 21Volts rail. 😉
current at VAS line is about 8,3mA at R20 (degen. resistor Q5)...okay
if i measure from GND to the base the pnp (BDW94C) is more open --> 1,46V, NPN is just 1,04V

have to check the values maybe tomorrow.

have a nice evening
chris

 

[de Ocampo]

Nice work Chris, dc offset can be significantly reduced by matching the input pair. If that didn't work you may have to reduce the values of the degenerators, I guess you are using too high a value for a slower devices (2n5401). I am judging from the schematic in post #97.
My experience with simulation, a dc offset of about 1mv gives you a much closer result in the real world build.

Albert

edit: I was late to realize that you are using TO-126 for the VAS.

 

[chermann]

i did a sim if is use the real value for the voltage of the red LED.(1,74V) its easy to find a diode.
i have to change the R3 (CCS) to 520ohm to get my favorite 2,1mA. then if to change the R6 to about 777ohms to get a perfect balance at my amp. DC offset is then 6,7mV.

 

[de Ocampo]

I'll see if I can make the .asc file work, I once tried to sim a darlington but I get errors.

 

[chermann]

here is the zip with my last changes

 

[chermann]

Hi
short update
i tireid to get the values for the resistors but its not easy.
finaly i have got 35mV DC offset after 20min and 50mA quiecent current.

interessting is that with 21,5V i got DC offset about 4mV. so i think i could be that a 15VA transformer could give me about 21,5V because the "load" is not so much with 2 amps.

kr
chris

 

[chermann]

Hi
i just let run at idle current of 106mA / 21V rail since 6 hours and wverything is fine. DC offset is about -33mV. i let run the scope to look at some graphs.
current is now stable also the DC offset is not ideal but stable.
no heat - everything is less then luke warm 😉

kr
chris

 

[chermann]

hi
here is a first test with 9R and just to small 5W/18R resistors parallel.
this amp is stable with 10kHZ and 100nf and 390nF at about 5WATT

sine wave i test is done with about 7,4WATT and the upper fg was with 9R load 280Khz

not bad 😉

kr
chris

 

[chermann]

Hi

here are some measurments with my smal load resistor. the values can be read in the filename

 

[chermann]

any hints what i should be looking for?
i will do the load test with cap again...

 

[chermann]

square wave at 9R and about 9WATT with cap load.

 

[chermann]

on_off pop with 21Volt rail is okay

 

[chermann]

thx mooly 😉

 

[chermann]

hi
some measurements at 4R and 8R with my L channel.

still DC ofset with 21Vrail too high for my taste.
at home i´v got -48mvDC offset...

tests are fine.