DC offset and DC Protection question

[captJackSparrow]

I have limited knowledge about electronics and still learning

DC offset is presence of DC at output which is bad if it is too high, after reading many posts anything like 50mv DC should be alright. I understand that a low DC offset is not going to kill any driver.

I was reading about Speaker protection and speaker failures(google), in case full DC voltage is present at output.
-Relays do not provide 100% protection.
-Somewhere it was mentioned to use bipolar cap between Amp O/P and Speaker I/P to block any DC, it was mentioned that a 10000microfarad cap will have not affect on O/P AC signal.

Question
1. Do bipolar caps blocks DC and provides 100% protection from full DC voltage at output?

2. If answer to #2 is Yes, do they have any affect on AC O/P signal from AMP?
3. If answer to #2 is No, then why doesn't everyone uses it to completely block DC from O/P?


My Primary concern/requirement/query is 100% DC Protection due to MOSFET failure or any other reason.

 

[JonSnell Electronic]

1, Bipolar caps do provide DC isolation.
2. They do have an effect on AC signal. You can work out the 3db point using F=1/2pi R C
3. They are costly and a DC offset is normally controllable.

 

[El Reino]

Question 3, some people hate caps.

 

[jan.didden]

Big caps between the amp output and the speaker are normally only used (and required) in amps with a single power supply voltage. In such amps the output sits at half the supply voltage, hence you need the cap.
In a sense those amps have an offset of half the supply voltage ;-)

Jan

 

[captJackSparrow]

1, Bipolar caps do provide DC isolation.
2. They do have an effect on AC signal. You can work out the 3db point using F=1/2pi R C
3. They are costly and a DC offset is normally controllable.

Thankyou,

I don't understand the second part of #2 yet. What is the impact in terms of sound.
Regarding #3, a bipolar cap can be made by connecting two electrolytic 4700microFarad caps.

Question 3, some people hate caps.

Thankyou.

Big caps between the amp output and the speaker are normally only used (and required) in amps with a single power supply voltage. In such amps the output sits at half the supply voltage, hence you need the cap.
In a sense those amps have an offset of half the supply voltage ;-)

Jan

Thankyou, i can't afford a blown driver so i have to use caps for DC isolation unless it has huge impact on sound quality as i have read that relays can let DC through.

 

[juanitox]

4)
change your amp for a transformer coupled one or with an output cap like many of PAss Zen or SIT amp .

 

[captJackSparrow]

4)
change your amp for a transformer coupled one or with an output cap like many of PAss Zen or SIT amp .

Thankyou

I am already set on F5V2(70% board stuffed).

 

[jan.didden]

You should realize that the majority of amps out there have bipolar power supplies and some kind of DC output control and are fine. A blown speaker due to DC offset is extremely rare - actually never heard about it but I am only into audio 50 years.

Think about it, even if the offset would be a whopping one volt, that's still not going to blow your 100W speaker. 1V on a 8 ohms speaker is only 125mW ...

Jan

 

[synonymous]

I'm with Jan

I use a primitive DC protection circuit that kills the AC mains power to the amp
Resistor, N/P Electrolytic, Optoisolator, Latching relay controlling mains relay
7 parts total
No speaker relays
Probably never ever activate beyond testing, ever

 

[nigelwright7557]

I have made a few speaker protect pcb's for myself.
From simple DC detect and drop relay to a full DC protect, over current detect and power up/ down relay inhibit.
I used to use normal relays but now only use solid state relays as they behave better with switching DC. They are more expensive than a cheap mechanical relay but not too much more.
I tend to use small PIC microcontrollers to get some intelligence in there.

 

[MEPER]

Both ACA and MoFo amps has output caps so maybe something to consider?
Which amp do you consider?
The M2X has no output caps so has a very small about of DC-offset but seems very safe in case of failure. For most speakers only the woofer "see" DC from amp and woofer can take some amount of DC without burning. The other speaker units are in most cases projected by a cap in crossover filter. So you could check your speakers and if only woofer gets DC from amp you could protect only the woofer with a cap.

 

[nigelwright7557]

Capacitors can age and eventually go short circuit.

 

[jan.didden]

Both ACA and MoFo amps has output caps .

If memory serves, these are single supply amps so the output sits at half supply. Of course you need an output cap!

(This is a repeat from my earlier post which you probably didn't read).

Jan

 

[MEPER]

The mention of ACA and MoFo was just a suggestion of amps to build where an output capacitor is integrated in the design. MoFo can run without an output cap if DC resistance of choke is low enough.

 

[captJackSparrow]

You should realize that the majority of amps out there have bipolar power supplies and some kind of DC output control and are fine. A blown speaker due to DC offset is extremely rare - actually never heard about it but I am only into audio 50 years.

Think about it, even if the offset would be a whopping one volt, that's still not going to blow your 100W speaker. 1V on a 8 ohms speaker is only 125mW ...

Jan

"A blown speaker due to DC offset is extremely rare - actually never heard about it but I am only into audio 50 years."

My primary concern is 100% DC Protection due to MOSFET or any other failure which could lead to full DC voltage at output. DC Offset was just a side query that if i use bipolar caps at output, DC offset would be taken care of. Sorry for not being clear.


So my question is, does adding bipolar caps at output provide 100% DC Protection without impacting sound adversly, any known reliable way to get DC protection other than Bipolar Caps and relays. As i have read relays in protection circuit arc and allow full DC voltage at O/P.

I'm with Jan

I use a primitive DC protection circuit that kills the AC mains power to the amp
Resistor, N/P Electrolytic, Optoisolator, Latching relay controlling mains relay
7 parts total
No speaker relays
Probably never ever activate beyond testing, ever

Can you please share your DC Protection circuit?
Also you mentioned N/P Electrolytic - You mean you are using NonPolar cap at Output?
In your protection circuit, say you get full DC voltage at output, is it possible that relay may fail to protect, will other components(N/P Electrolytic, Optoisolator) still be in the path and protect from DC ?

I have made a few speaker protect pcb's for myself.
From simple DC detect and drop relay to a full DC protect, over current detect and power up/ down relay inhibit.
I used to use normal relays but now only use solid state relays as they behave better with switching DC. They are more expensive than a cheap mechanical relay but not too much more.
I tend to use small PIC microcontrollers to get some intelligence in there.

Can you please share your DC Protection circuit, is Solid state relay prone to arc issue at full DC voltage at output?

Both ACA and MoFo amps has output caps so maybe something to consider?
Which amp do you consider?
The M2X has no output caps so has a very small about of DC-offset but seems very safe in case of failure. For most speakers only the woofer "see" DC from amp and woofer can take some amount of DC without burning. The other speaker units are in most cases projected by a cap in crossover filter. So you could check your speakers and if only woofer gets DC from amp you could protect only the woofer with a cap.

Agree, I am building F5 T V2

Capacitors can age and eventually go short circuit.

Thankyou for highlighting.

Thankyou all for your valuable feedback.

 

[MEPER]

Ok.....I missed that your target is F5 T.
An output cap will always impact the sound. It depends of the quality etc. Nothing is 100%.....but I don't assume that your life depends of not blowing the speakers? ….so a good cap will be safe enough. You can test sound with and without cap to find out if you can live with the cap.

 

[plasnu]

I wonder what's wrong with the relay protection. If the rail is 24V or so, short period of DC would really harm the speakers?

 

[nigelwright7557]

I wonder what's wrong with the relay protection. If the rail is 24V or so, short period of DC would really harm the speakers?

In the distant past I have seen me build a new amp and then have the output transistors fail and put out DC. What usually happens is the sound goes off except for severe hum through speaker. If you can get to power switch fast enough then maybe that's ok.
A good few years ago I decided to build a test bed for new amplifiers.
It has power supply and volume control on crocodile leads and a box on the output that detects DC and cuts a relay if a problem occurs. Must have saved me a fortune in speakers over the years. Its just a PIC micro with a relay, not exactly rocket science.

 

[nigelwright7557]

Can you please share your DC Protection circuit?
Can you please share your DC Protection circuit, is Solid state relay prone to arc issue at full DC voltage at output?

The circuit isn't much use to you as it partly software based.
It uses a PIC micro with a solid state relay.
Sadly it needs a power supply which is a nuisance.
The beauty of the PIC is you can add things like a power up delay.
My all singing all dancing unit also does power down and over current detect.
Although it is quite a large pcb.

Mosfet's don't arc. Although, it does need a small cap across it to dampen spikes.

 

[captJackSparrow]

The circuit isn't much use to you as it partly software based.
It uses a PIC micro with a solid state relay.
Sadly it needs a power supply which is a nuisance.
The beauty of the PIC is you can add things like a power up delay.
My all singing all dancing unit also does power down and over current detect.
Although it is quite a large pcb.

Mosfet's don't arc. Although, it does need a small cap across it to dampen spikes.

Thankyou
Since yesterday I digged further, I believe MOSFET/Optocoupler is the way to go unless somes highlights any serious flaw like with mechanical relays.

I have seen many circuits on this forum, couldn't find any standard one which has been tested for full DC voltage. No clear confirmatiin that the circuit shown is final circuit, I can copy circuits but I can't alter/add anything on my own as I dont possess that kind of knowledge. Most of them are theory/suggestions. If you know of any simple circuit using optocoupler and mosfet with very fast response please share, ill keep searching the forum.