Sziklai power amplifier

maouna · 9th March 2010, 09:19 PM

what is your opinion about this amp??
driver and vas stage transistors are 2SD669A,2SB649A

Joachim Gerhard · 9th March 2010, 09:35 PM

looks like a D.Self design.

nigelwright7557 · 9th March 2010, 09:37 PM

You might find Q70 needs a 100pf from collector t obase to stop oscilation on the output. I certainly found that in my quasi amp.

maouna · 9th March 2010, 10:01 PM

yes its based on d.self's book. i have made this amp and now iam testing it but i am having a problem.
when i power it on ,theres a constant current of 500ma at one rail,regardless the input..

sometimes when i power it on,its ok but after turning up the volume a bit ,theres this constant current of 500ma .
whats this all about?

mlloyd1 · 9th March 2010, 10:06 PM

which rail?
how much bias are you using?
do you have a scope to look at the output?

mlloyd1

Rikard Nilsson · 9th March 2010, 10:08 PM

Quote:

Originally Posted by maouna

yes its based on d.self's book. i have made this amp and now iam testing it but i am having a problem.
when i power it on ,theres a constant current of 500ma at one rail,regardless the input..

sometimes when i power it on,its ok but after turning up the volume a bit ,theres this constant current of 500ma .
whats this all about?

I would strongly suspect some kind of oscillation that triggers on a bit of output power (unless you get DC offset that draws current from one rail only). Once the oscillation starts it might not stop by itself. The amp needs stabilization. These kinds of output stages are prone to oscillation problems, so nothing surprising really. Emitter follower outputs are easier to stabilize, eventhough I haven´t succeeded completely myself in my own design.

cbdb · 9th March 2010, 10:10 PM

Is it oscillating? You will need a scope to find out, because the oscillations could be in the megahertz. Try the caps Nigel sugested.

maouna · 9th March 2010, 10:23 PM

thanks all for your advice. its the negative rail i have connected the ammeter to,.i dont have a scope unfortunatelly..

dc offset is about 5-6 mV, and bias current is about 25ma.

the sound is ok at the output ,just this problem with the current.

the zobel resistor doesnt get hot .i suspect that when an amp is oscillating ,it drains alot of current but in my case its just 500-600ma

krachkiste · 11:07 PM

Unfortunately you will not get very far in finding out what happens without a scope.

Try reducing R41 in steps down to 220R.

After that, try the folloing:
Try adding in series to Q67 emitter a parallel-circuit of the following three elements (don't leave one out, they're all important):
- a resistor smaller or equal to 68R (in steps, R_max=0.6V/I_q_VAS) in parallel with
- a diode (e.g. 1n4148) (kathode tied to rail anode at emitter of Q67 to prevent the emitter of rising above 0.7V over rail voltage) and also in parallel
- a capacitor of about 470pF to keep the gain on track at HF.

instead of a diode you could also use a transistor as self did in his book for current limiting. (guess that might be even more effective).

Try to measure the voltage over the emitter-Rs (R14 + R17) from emitter to emitter.
Check if there are jumps or hystheresis in the voltage displayed on your voltmeter, while setting the bias (no input signal). If the voltage is zero but the current through Q55,Q68,Q57 and Q70 is not -> pretty sure oscillation.

jaycee · 9th March 2010, 11:14 PM

The PCB layout may also be a factor. Sziklai's are quite sensitive to this.

Your schematic doesnt show any on-PCB power supply decoupling. You will want this. Put some 220-330uF decoupling capacitors on the PCB, near to the output transistors. Bypass with 100nF.

9th March 2010, 09:37 PM	#3
nigelwright7557 diyAudio Member Join Date: Apr 2008 Location: Carlisle, England	You might find Q70 needs a 100pf from collector t obase to stop oscilation on the output. I certainly found that in my quasi amp. __________________ http://www.murtonpikesystems.co.uk PCBCAD40 pcb design software.

9th March 2010, 11:00 PM	#9
krachkiste diyAudio Member Join Date: May 2009	Unfortunately you will not get very far in finding out what happens without a scope. Try reducing R41 in steps down to 220R. After that, try the folloing: Try adding in series to Q67 emitter a parallel-circuit of the following three elements (don't leave one out, they're all important): - a resistor smaller or equal to 68R (in steps, R_max=0.6V/I_q_VAS) in parallel with - a diode (e.g. 1n4148) (kathode tied to rail anode at emitter of Q67 to prevent the emitter of rising above 0.7V over rail voltage) and also in parallel - a capacitor of about 470pF to keep the gain on track at HF. instead of a diode you could also use a transistor as self did in his book for current limiting. (guess that might be even more effective). Try to measure the voltage over the emitter-Rs (R14 + R17) from emitter to emitter. Check if there are jumps or hystheresis in the voltage displayed on your voltmeter, while setting the bias (no input signal). If the voltage is zero but the current through Q55,Q68,Q57 and Q70 is not -> pretty sure oscillation. Last edited by krachkiste; 9th March 2010 at 11:07 PM.

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
a wonderfull sziklai	sakis	Solid State	198	25th January 2012 10:29 PM
Negative Sziklai	timH	Solid State	17	30th July 2007 11:13 PM
DC adjustable double-Sziklai Power Diamond Buffer	mtlin12	Solid State	22	16th July 2005 07:11 AM
sziklai-darlington diamond buffer (SDDB) for power output stages	mtlin12	Solid State	5	1st April 2005 07:55 AM

New To Site?	Need Help?
Register to Participate Search Privacy Statement Contact Us	Frequently Asked Questions Did you forget your password? Mark Forums Read


	Please consider donating to help us continue to serve you. Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving

9th March 2010, 09:35 PM	#2
Joachim Gerhard diyAudio Member Join Date: Mar 2009	looks like a D.Self design.

9th March 2010, 10:01 PM	#4
maouna diyAudio Member Join Date: Oct 2008	yes its based on d.self's book. i have made this amp and now iam testing it but i am having a problem. when i power it on ,theres a constant current of 500ma at one rail,regardless the input.. sometimes when i power it on,its ok but after turning up the volume a bit ,theres this constant current of 500ma . whats this all about?

9th March 2010, 10:06 PM	#5
mlloyd1 diyAudio Member Join Date: Feb 2001 Location: illinois	which rail? how much bias are you using? do you have a scope to look at the output? mlloyd1

9th March 2010, 10:10 PM	#7
cbdb diyAudio Member Join Date: Oct 2008 Location: Vancouver	Is it oscillating? You will need a scope to find out, because the oscillations could be in the megahertz. Try the caps Nigel sugested.

9th March 2010, 10:23 PM	#8
maouna diyAudio Member Join Date: Oct 2008	thanks all for your advice. its the negative rail i have connected the ammeter to,.i dont have a scope unfortunatelly.. dc offset is about 5-6 mV, and bias current is about 25ma. the sound is ok at the output ,just this problem with the current. the zobel resistor doesnt get hot .i suspect that when an amp is oscillating ,it drains alot of current but in my case its just 500-600ma

9th March 2010, 11:14 PM	#10
jaycee diyAudio Member Join Date: Apr 2004 Location: Norwich, UK	The PCB layout may also be a factor. Sziklai's are quite sensitive to this. Your schematic doesnt show any on-PCB power supply decoupling. You will want this. Put some 220-330uF decoupling capacitors on the PCB, near to the output transistors. Bypass with 100nF.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)