amplifier attempt #1

name773 · 5th July 2005, 09:44 PM

here's an amplifier i just designed with great assistance from Rod Elliot's articles and a select few of his designs. It is intended to drive a pair of grado sr80s, any ideas if it will sound ok/be stable/etc? thank you for any shared insight

Fritzell · 5th July 2005, 10:18 PM

Looks sweet. But im thinking some thermal resistors could come in handy on the emitter of bd139/140.. then youre very close to a standard amp. Just change the input stage before the VAS to a differential input. And maybe there would be good to have a dominant capacitor to erase oscillation. Also a capacitor series with resistor on the output ?

darkfenriz · 5th July 2005, 10:23 PM

Hi
this baby seems to have unity gain, create a V divider with a resistor from output to emitter of Q1.
You can also give some smallish (like 0R33) emitter resistors for output transistors and create a 'diamond' at input (add a npn emitter follower) to minimize offset voltage.
Just a first glance, might be wrong.
best regards

name773 · 5th July 2005, 10:47 PM

oops, i forgot the other resistor that determines the gain...

i would like to stay away from the differential input to preserve a smallish component count (i hope to make this thing portable and stereo)
unless of course, that would help minimize voltage offset. i'd rather not add another transistor between the input transistor and the output driver, because the collector current on the input transistor should be around 120uA to drive the base of the driver transistor, and that's pretty low as it is.

would it be ok to make R4 1k8 and C2 10uf? that gives a roll off frequency of ~8.84 hz, but will it allow enough current through?

Eva · 5th July 2005, 10:48 PM

I think that the input transistor is not properly biased. The DC output of the circuit will be more than one diode drop below ground, so an output coupling capacitor may be required.

PRR · 5th July 2005, 11:02 PM

The output DC voltage will be around -0.6VDC, which will not do the Grados any good. Either add a second input device to give near-zero offset, or re-cast the whole thing with +12V supply and capacitor coupling input and output.

There is no shorted-load protection. Headphone cables get pinched, Q2 and Q6 try to pass infinite current. Your second iteration has 0.33 ohm resistors, cutting shorted-load current to only 18 Amps, a little more than Q6 is rated for and far more heat than you can handle.

I'm glad to see R10 R11 added (how would you adjust bias without a resistor somewhere to stick your voltmeter?) but still think they are too low. 10% of load impedance is a better goal. A more subtle goal (for Class AB operation) is 30mV-100mV at idle current. Assuming AB idle bias is 1/10th of peak current, both "rules" lead to about 3 ohms, not 0.3 ohms.

darkfenriz · 5th July 2005, 11:04 PM

Now it looks way better. I think bigger R4 may play even better, but remember to make R9 bigger too. In project this simple the best thing to do is build and experiment a bit, maybe some simulation first. Try doing it, don't let any one to put you off.

name773 · 6th July 2005, 12:40 AM

thank you all for the excellent replies!

here is an updated version

darkfenriz · 6th July 2005, 01:01 AM

shouldn't base of q5 be to the right 'leg'?
Now you have much more amplifying devices and more open loop gain. Give q1 q2 q3 q4 and q6 emitter resistors of value somewhere in range 47R-330R for better performance and stability.

amplifierguru · 6th July 2005, 01:10 AM

Yes Darkfenriz,

Other leg for Q5 base and all the resistors stated. Q6 emitter should be just below R7 in value so keep the others small.

Cheers,
Greg

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
My first attempt	mikee55	Chip Amps	4	25th March 2006 11:16 PM
Amplifier Attempt 2	scott_jcp	Solid State	10	11th April 2005 12:13 AM
First amplifier attempt	scott_jcp	Solid State	7	7th April 2005 11:35 PM

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

5th July 2005, 10:18 PM	#2
Fritzell diyAudio Member Join Date: Jul 2004 Location: Sweden	Looks sweet. But im thinking some thermal resistors could come in handy on the emitter of bd139/140.. then youre very close to a standard amp. Just change the input stage before the VAS to a differential input. And maybe there would be good to have a dominant capacitor to erase oscillation. Also a capacitor series with resistor on the output ?

5th July 2005, 10:23 PM	#3
darkfenriz diyAudio Member Join Date: Jun 2004 Location: Warsaw	Hi this baby seems to have unity gain, create a V divider with a resistor from output to emitter of Q1. You can also give some smallish (like 0R33) emitter resistors for output transistors and create a 'diamond' at input (add a npn emitter follower) to minimize offset voltage. Just a first glance, might be wrong. best regards

5th July 2005, 10:48 PM	#5
Eva diyAudio Member Join Date: Oct 2003 Location: Near the sea	I think that the input transistor is not properly biased. The DC output of the circuit will be more than one diode drop below ground, so an output coupling capacitor may be required.

5th July 2005, 11:02 PM	#6
PRR diyAudio Member Join Date: Jun 2003 Location: USA	The output DC voltage will be around -0.6VDC, which will not do the Grados any good. Either add a second input device to give near-zero offset, or re-cast the whole thing with +12V supply and capacitor coupling input and output. There is no shorted-load protection. Headphone cables get pinched, Q2 and Q6 try to pass infinite current. Your second iteration has 0.33 ohm resistors, cutting shorted-load current to only 18 Amps, a little more than Q6 is rated for and far more heat than you can handle. I'm glad to see R10 R11 added (how would you adjust bias without a resistor somewhere to stick your voltmeter?) but still think they are too low. 10% of load impedance is a better goal. A more subtle goal (for Class AB operation) is 30mV-100mV at idle current. Assuming AB idle bias is 1/10th of peak current, both "rules" lead to about 3 ohms, not 0.3 ohms.

5th July 2005, 11:04 PM	#7
darkfenriz diyAudio Member Join Date: Jun 2004 Location: Warsaw	Now it looks way better. I think bigger R4 may play even better, but remember to make R9 bigger too. In project this simple the best thing to do is build and experiment a bit, maybe some simulation first. Try doing it, don't let any one to put you off.

6th July 2005, 01:01 AM	#9
darkfenriz diyAudio Member Join Date: Jun 2004 Location: Warsaw	shouldn't base of q5 be to the right 'leg'? Now you have much more amplifying devices and more open loop gain. Give q1 q2 q3 q4 and q6 emitter resistors of value somewhere in range 47R-330R for better performance and stability.

6th July 2005, 01:10 AM	#10
amplifierguru Account Disabled Join Date: Aug 2004 Location: Earth	Yes Darkfenriz, Other leg for Q5 base and all the resistors stated. Q6 emitter should be just below R7 in value so keep the others small. Cheers, Greg