What is DC offset?

spooney · 29th February 2008, 12:06 PM

I've read many threads on here about repairs that talk about dc offset but I am still confused as to what exactly dc offset is and what it means as far as amp repairs.It would be greatly appreciated if somebody could shed some light on the subject for me.

richie00boy · 29th February 2008, 12:11 PM

Its a DC voltage at the amp output. Typically if it's over 100mV that's not very good.

Perry Babin · 29th February 2008, 12:19 PM

Ideally, there will be no DC voltage (offset voltage) across the speaker terminals of an amplifier. Most amplifiers have a small amount of DC. Most people (those building or repairing amplifiers) consider 0.025v or less to be acceptable. 0.1v may be acceptable for others. Most ampliifers in good working order will have significantly less than 0.1v of DC offset.

Generally, when an amplifier fails and has DC offset, the DC voltage across the speaker terminals is approximately equal to the rail voltage.

1moreamp · 29th February 2008, 07:26 PM

I look and test for it. I also try to repair its causes. I like to see Like new spec's for this tech spec. I use the DC offset figure as a gauge of amp abuse, and shortened life span. Sorta like a gauge as to a amps overall health indication.

Just goggle DC offset and you will see others info posts about it, and its links to THD, and noise figures. I feel its a important thing to look at, But this is only my opinion,,,

spooney · 29th February 2008, 08:10 PM

what are some things that can cause the dc offset to rise?

1moreamp · 29th February 2008, 08:39 PM

Diff input pair damage, I find DC offset here, and its thermally enhanced by unequal temperatures of this group of transistors. Leaky outputs and drivers, bad caps. Resistors that have overheated and altered value especially in the input section. Leaky diode also can be a fault for this issue. I have also seem power supply imbalance picked up and sent to the output as DC offset due to bad filter caps in older amps.

PPI used a simple two resistor with a small value pot in the middle across the 15 volt rails to make a low level DC signal to insert in the invert input so any normal offset could be Nulled out at the input stage. I find this a simple method to the possible mismatch and or aged components of the amp circuitry.

All in all its a sure sign of many things. As most amps today are balanced complimentary design and with that comes DC offset related issues when the semi's have been aged by heat, abuse, and just plain old age and use. Nothing lasts forever, if it did we would only buy one of anything, and that would hinder world commerce....Especially Chinese imports

jol50 · 29th February 2008, 09:20 PM

If that does not make sense, the job of the amp is to make the speaker go in/out with AC current....+ voltage one way and - voltage the other. Sitting there idle with no input it should have 0 current to the speaker AC and never have DC....but few amps are perfect and a little makes no difference. Ones I see usually are under 50mv, better ones under 30 or 10mv even. Some amps are 0 depending on how they are made. I have one with 250mv I am trying to fix, if I would get to work in it again and put the new caps in

but not sure what is wrong.

1moreamp · 1st March 2008, 01:02 AM

with experience you will see what will pass for a reasonable offset spec. Many class D amps have like 250 to 500 MVDC offset at idle.

I have seen many new amps fresh out the box with 25 MVDC offset. But this was a imbalance in the power supply causing this, as when bridged the offset was 3 MVDC, so it was common to all channels and was inverted on the bridge channel so it canceled when bridged.

I generally shoot for the lowest possible offset figures, but sometimes that requires way too much time to be worthwhile so I generally talk to the customer about these, so a consensus can be drawn on the amp value, as opposed to repair cost.

On PPI's I generally am able to reduce the offset by adjustment to less than 1 MVDC without issue. Just a few minutes to adjust and is easily verified on burn-in final test.

Lower is better...

spooney · 1st March 2008, 01:29 PM

I have an amp on the bench right now that has .003 volts on the channel I just repaired and .027 volts on the channel that was original to the amp.Is this difference going to cause any major issues?

richie00boy · 1st March 2008, 04:09 PM

None whatsoever. It's not the difference that matters really unless you are running it bridged.

29th February 2008, 12:06 PM	#1
spooney diyAudio Member Join Date: Dec 2006	What is DC offset? I've read many threads on here about repairs that talk about dc offset but I am still confused as to what exactly dc offset is and what it means as far as amp repairs.It would be greatly appreciated if somebody could shed some light on the subject for me.

29th February 2008, 12:11 PM	#2
richie00boy Did it Himself diyAudio Member Join Date: Nov 2003 Location: Gloucestershire, England, UK	Its a DC voltage at the amp output. Typically if it's over 100mV that's not very good. __________________ And yes, there are capacitors in the circuit. One is even employed to form the dreaded bootstrap on the voltage gain stage. Get over it. -Burning Amplifier #2, Nelson Pass www.readresearch.co.uk my website for UK diy audio people - designs, PCBs, kits and more

29th February 2008, 12:19 PM	#3
Perry Babin diyAudio Member Join Date: Nov 2003 Location: Louisiana	Ideally, there will be no DC voltage (offset voltage) across the speaker terminals of an amplifier. Most amplifiers have a small amount of DC. Most people (those building or repairing amplifiers) consider 0.025v or less to be acceptable. 0.1v may be acceptable for others. Most ampliifers in good working order will have significantly less than 0.1v of DC offset. Generally, when an amplifier fails and has DC offset, the DC voltage across the speaker terminals is approximately equal to the rail voltage. __________________ Basic Car Audio Amp Repair Basic Car Audio Electronics Basic Transistor Testing

1st March 2008, 04:09 PM	#10
richie00boy Did it Himself diyAudio Member Join Date: Nov 2003 Location: Gloucestershire, England, UK	None whatsoever. It's not the difference that matters really unless you are running it bridged. __________________ And yes, there are capacitors in the circuit. One is even employed to form the dreaded bootstrap on the voltage gain stage. Get over it. -Burning Amplifier #2, Nelson Pass www.readresearch.co.uk my website for UK diy audio people - designs, PCBs, kits and more

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
DC Offset	JannaCassandra	Chip Amps	3	26th July 2008 04:20 PM
2,5v dc offset!!	nicoch46	Class D	2	5th March 2007 12:02 AM
Help With DC Offset	Sybex23	Chip Amps	8	15th November 2004 08:11 PM
GC offset	janneman	Chip Amps	4	3rd July 2004 09:10 AM
GC very low DC offset	matjans	Chip Amps	7	22nd September 2003 11:24 PM


<a rel='nofollow' href='http://www.diyaudio.com/openx/www/delivery/ck.php?n=aa5cb2e9&cb=623596' target='_blank'><img src='http://www.diyaudio.com/openx/www/delivery/avw.php?zoneid=3&cb=623596&n=aa5cb2e9' border='0' alt='' /></a>	We're saving for a new server - help us to serve you by Donating Today and become a friend with benefits! Ads on/off / Custom Title / 2009 Tshirt / More PMs / Bigger Images / Advanced printing

29th February 2008, 07:26 PM	#4
1moreamp diyAudio Member Join Date: Dec 2005 Location: Northern California	I look and test for it. I also try to repair its causes. I like to see Like new spec's for this tech spec. I use the DC offset figure as a gauge of amp abuse, and shortened life span. Sorta like a gauge as to a amps overall health indication. Just goggle DC offset and you will see others info posts about it, and its links to THD, and noise figures. I feel its a important thing to look at, But this is only my opinion,,,

29th February 2008, 08:10 PM	#5
spooney diyAudio Member Join Date: Dec 2006	what are some things that can cause the dc offset to rise?

29th February 2008, 08:39 PM	#6
1moreamp diyAudio Member Join Date: Dec 2005 Location: Northern California	Diff input pair damage, I find DC offset here, and its thermally enhanced by unequal temperatures of this group of transistors. Leaky outputs and drivers, bad caps. Resistors that have overheated and altered value especially in the input section. Leaky diode also can be a fault for this issue. I have also seem power supply imbalance picked up and sent to the output as DC offset due to bad filter caps in older amps. PPI used a simple two resistor with a small value pot in the middle across the 15 volt rails to make a low level DC signal to insert in the invert input so any normal offset could be Nulled out at the input stage. I find this a simple method to the possible mismatch and or aged components of the amp circuitry. All in all its a sure sign of many things. As most amps today are balanced complimentary design and with that comes DC offset related issues when the semi's have been aged by heat, abuse, and just plain old age and use. Nothing lasts forever, if it did we would only buy one of anything, and that would hinder world commerce....Especially Chinese imports

29th February 2008, 09:20 PM	#7
jol50 diyAudio Member Join Date: Nov 2007	If that does not make sense, the job of the amp is to make the speaker go in/out with AC current....+ voltage one way and - voltage the other. Sitting there idle with no input it should have 0 current to the speaker AC and never have DC....but few amps are perfect and a little makes no difference. Ones I see usually are under 50mv, better ones under 30 or 10mv even. Some amps are 0 depending on how they are made. I have one with 250mv I am trying to fix, if I would get to work in it again and put the new caps in but not sure what is wrong.

1st March 2008, 01:02 AM	#8
1moreamp diyAudio Member Join Date: Dec 2005 Location: Northern California	with experience you will see what will pass for a reasonable offset spec. Many class D amps have like 250 to 500 MVDC offset at idle. I have seen many new amps fresh out the box with 25 MVDC offset. But this was a imbalance in the power supply causing this, as when bridged the offset was 3 MVDC, so it was common to all channels and was inverted on the bridge channel so it canceled when bridged. I generally shoot for the lowest possible offset figures, but sometimes that requires way too much time to be worthwhile so I generally talk to the customer about these, so a consensus can be drawn on the amp value, as opposed to repair cost. On PPI's I generally am able to reduce the offset by adjustment to less than 1 MVDC without issue. Just a few minutes to adjust and is easily verified on burn-in final test. Lower is better...

1st March 2008, 01:29 PM	#9
spooney diyAudio Member Join Date: Dec 2006	I have an amp on the bench right now that has .003 volts on the channel I just repaired and .027 volts on the channel that was original to the amp.Is this difference going to cause any major issues?

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

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