Attachments
1.Then try change C23 with 1,5nF and R49 with 180-220R
2. Add 15-22pF from Base Q6 to collector Q8
3.Increase C8 or C22 to 10uF/16V+
4.Increase C10 and C11 to 150pF
BUT First-of-All
recalculate or measure voltage drop on R27 and R28 so we`ll find VAS bias on those 100R resistors!
LTP bias =(15V/5K/2)=3mA/2 = 1,5mA per BJT in input diff.pair = OK
R16, R17 voltage drop = 1,5mA*3K3 =4,95--->5V
VAS bias = (5V-Vbe_Q5)/100R = 4,3/100=43mA??? not OK!!!
dissipation of caskode Q8 and Q7
will be then aprocs. 50V*43mA=more then 2W , this ends as very-very hot VAS bias
Decrease value for both R16 and R15 from 3K3 to 910R-1,2K
so we get then R27, R28 bias close to 12-10mA
with dissipation of Q8 and Q7 and Q9 now: P_diss=50*10mA=>0,5-0,6W
diodes strings D5+D5, D8+D9, D10+D11 will work with 1-1,2mA bias
so we ends with VAS bias <10mA
Now first turn trimmpot R18 to max value so we get at start minimum BIAS,
Short input to SignalGND
Put Dummy load 4-10R (several W) on output
Power up the amp with limiter resistors (15-20R, several W) on rails,
then readjust R18 to read on DMM voltage drop on R39 (or on another output BJT emmiter resistors) those racomanded 22-25mV
= 23,5mV/0,33R = 70-75mA BIAS per output BJT pair
readjust bias after 10-15min amp idle runing.
Measure DC offset at output : +/-10mV OK
Control temperature of Main Heatsink, VAS Caskode BJTs Q7, Q8, Q9
Last edited:
Sunny22g,
it's your willingness to learn would decide. there are a lot of people who are willing to help you grow, you just have to understand and perform their advises- repeat it several times if necessary so you can learn.
Like you i have ask so many questions before but some of those questions had been answered in the past already so may i suggest to read the forum from the start as i did.
about 3 years ago i have started electronics from just googling ideas about amps my first amp was from Columbian websites - it works great, then made leach amps still it work. and now that i have found APEX by Mr. Mile its not just great when we talk about QUALITY and POWER.
i have made a lot of his designs with all has astonishing results. I have started with B500 next was AX14, FH9, SR200 with reversed version, SR250, AX14 3&4pairs, H900, pre-amps. active and passive crossovers and even speaker boxes.
and these people who helped you are the same people who helped me namely Sir Mile, Sir Dragan100, *AndrewT*, vzaichenko and a lot more...
to the D.I.Y. community THANK YOU for the help and support to the beginners i may not become expert nor earn profit but i have fulfilled my Hobby...
God Bless us all... ... ...
Hello Sir , please tell me where I got R49. I cannot see this. Plz tell me thanks Sry for bad english
Lead and lag comp. on this topology ??
Dragon was right , but 33-47pf from
C Q8 to B Q6 - just simple miller comp.
Faster Q8 and 9 , not high cob slow mje340-350 - maybe even C4793/A1837.
Get rid of the 18pf/ and R18-C7
"abomination".
OP stage - huge 220R base-stoppers on the drivers ? Gone , add 5-10R base-stoppers to all the output devices.
With above alterations , <1mhz unity gain with @ 80 degrees phase margin can be
expected. (super stable with even enough derating to allow "junk drawer" semi's ).
Not to denigrate the design , but it ignores many of the audio design rules
laid out in both the Self and Cordell books ???
OS
Last edited:
I dropped my dream to make apex sr200. But droping sr200 . because there are many problems.... aur there are no diy member in my city.who can help my problems . I really love this amplifier because my friend have studio monitor . sound quality is very imba but I cannot full fill my dream s.
Last edited:
Hello I am totally confused.
Remove C7 (470pF), connect 47pf from C Q8 to B Q6, do not use 18pF (I can not see it on your build)...
Attachments
Sr200
Sir kandimba I will uploading 3.2 Ver .sir Juan give this zip there is PCB pictures.plz tell me where r49.
I sunny, you should see this diagram. And be cool)
Sir kandimba I will uploading 3.2 Ver .sir Juan give this zip there is PCB pictures.plz tell me where r49.
