EDAboard.com | EDAboard.de | EDAboard.co.uk | WTWH Media

Speaker Protection Question.

Ask a question - edaboard.com

elektroda.net NewsGroups Forum Index - Electronics AUS - Speaker Protection Question.

Goto page Previous  1, 2, 3, 4  Next

Chris Jones
Guest

Thu Jun 29, 2017 3:43 pm   



On 29/06/2017 17:17, Phil Allison wrote:
Quote:
Chris Jones wrote:


FYI a link to Phil's post:
https://groups.google.com/d/msg/sci.electronics.repair/nwWqmrjKEsU/A1SGGVcCbj8J

Full Text:
** I have just been experimenting with a PCB mount, 240VAC/12amp relay
made by "Schrack" and a variable PSU - consisting of a 300VA tranny,
bridge rectifier and 10,000uf cap. The tranny was fed from a Variac.

The relay, when energised, connected a 4ohm high power load to the PSU
and tried to disconnect it when de-energised. With no cap across the
contacts, serious flash arcing occurred at 30VDC. With 50VDC, you can
normally expect the arc to become continuous, first shot. The maker's
rating for DC switching is 24V at 10 amps max.

However, with a 20uF film cap across the contacts, all signs of arcing
at switch off disappeared. Amazingly, this was still the case when tried
with 6, 3 & 1uF instead. When I tried 0.22uF, flash arcs appeared on
about 1 out of 3 tries.

To simulate a *bad* inductive speaker load, I added a 5mH air core choke
in series with the 4 ohm load and saw slight flash arcing with a 1uF cap
but none with 3uF.

With a 8 ohm load and 100VDC, 6uF was enough to reduce arcing to minor
flashes.

So, a film cap across the relay contacts made a huge difference when
breaking DC current at voltages well above the relay's ratings.

FYI: the film cap delays the voltage rise across the relay's contacts
for the first 10 to 50 *microseconds* after opening - which is when the
arc forms. Delay that rise enough and there is no arc.



** Yep, that was posted my me some time back.

A 10uF, 200V or better film cap would do the job very nicely.

But there is a problem:

Speaker relays are mainly used to STOP switch on noises - like loud cracks, squeals and thuds.

Having a 10uF cap across the contacts lets a lot of it through - certainly to the mids and tweeters in a system.


Yes, though in the present situation with no speaker protection at all,
it is just as bad, and adding the relay with capacitor might help the
speaker to survive an amplifier failure without burning out. Perhaps
this particular amplifier does not suffer from those switch on noises
anyway, e.g. due to lucky supply sequencing.

If the relay has two sets of contacts that are able to be connected in
series, then a capacitor could be connected across just one set of
contacts. The contacts with the capacitor would be able to switch off
the speaker hopefully without arcing in the event of an amplifier fault,
and the contacts with no capacitor would stop the clicks and pops if
they occur when the relay coil is not energised.

Phil Allison
Guest

Thu Jun 29, 2017 4:06 pm   



Chris Jones wrote:

-----------------------


Quote:

FYI a link to Phil's post:
https://groups.google.com/d/msg/sci.electronics.repair/nwWqmrjKEsU/A1SGGVcCbj8J


** Yep, that was posted my me some time back.

A 10uF, 200V or better film cap would do the job very nicely.

But there is a problem:

Speaker relays are mainly used to STOP switch on noises - like loud
cracks, squeals and thuds.

Having a 10uF cap across the contacts lets a lot of it through
- certainly to the mids and tweeters in a system.



Yes, though in the present situation with no speaker protection at all,
it is just as bad, and adding the relay with capacitor might help the
speaker to survive an amplifier failure without burning out.


** Using a DPDT relay wired like I suggested works a treat - and is totally silent when in the rest position.

It's the only method that deserves recommendation.



Quote:
Perhaps
this particular amplifier does not suffer from those switch on noises
anyway, e.g. due to lucky supply sequencing.


** When connected to a pre-amp, if both come on from the same AC switch - it almost certain to make a loud noise.


Quote:

If the relay has two sets of contacts that are able to be connected in
series,



** Tolerating one iffy relay contact in series with Hi-Fi speaker is bad enough thank you.




..... Phil

~misfit~
Guest

Sat Jul 15, 2017 6:15 am   



Once upon a time on usenet ~misfit~ wrote:
[snipped]
Quote:
The issue I have is that it has no speaker protection and my
speakers are irreplaceable (rarity and my budget considered). I have
two speaker protection PCBs sourced from AliExpress that look up to
the task, well designed with heavy duty relays and thick signal
traces made thicker by having the solder mask removed and solder
added. My problem is that I can't fit them and their power supply
into the amplifier case.

So I'm planning on putting them in a box between the amp and
speakers with speakers in and out connectors on the back. Just
checking to see if anybody thinks this isn't worth doing for any
reason? From what little I know they should work fine like this,
it's just a little fiddly.

These are apparently 30A but that can be taken with a pinch of salt;
https://www.aliexpress.com/item/upc1237-large-current-speaker-protection-board/32809522760.html
(I purchased what look like identical items from a different seller
who no longer has them for sale. I see that listing says maximum amp
20w - that seems as unlikey as the previous listings 600w.)


Well I finally got a transformer (EI as I didn't think a toroidal was needed
as the 'signal' is high-current) and built two of these single channel
protection PCBs into a re-puposed case. I figurd they're better than nothing
(if only for the built-in 3 second delay to avoid 'thump') until I can
rebuild a different two-channel upc1237 based PCB with remote relays to
short the speakers...

The damn thing introduces shitloads of hum! It took me a while to make back
and front panels for the case so only just got around to hooking it up with
my existing amps. (I prioritised this over finishing the refurb of the other
amp so I'd have /some/ protection ready and could use it when finished.) I
tried removing the transformer and running it oiutside the case but it still
hums.

Biugger! I don't get much low-pain 'mobile' time as it is after the
housework etc is done and I've just wasted a bunch building this thing.
Who'd have thought these would introduce so much hum? (Not me obviously.)
It's not as if I've got my stereo system laid out as I'd like it yet with
speaker wires and power cables separated AMAP and only crossing at
right-angles (I'm planning on putting it in a different place when it's all
done so...).
--
Shaun.

"Humans will have advanced a long, long way when religious belief has a cozy
little classification in the DSM*."
David Melville (in r.a.s.f1)
(*Diagnostic and Statistical Manual of Mental Disorders)

Trevor Wilson
Guest

Sat Jul 15, 2017 8:21 am   



On 15/07/2017 2:15 PM, ~misfit~ wrote:
Quote:
Once upon a time on usenet ~misfit~ wrote:
[snipped]
The issue I have is that it has no speaker protection and my
speakers are irreplaceable (rarity and my budget considered). I have
two speaker protection PCBs sourced from AliExpress that look up to
the task, well designed with heavy duty relays and thick signal
traces made thicker by having the solder mask removed and solder
added. My problem is that I can't fit them and their power supply
into the amplifier case.

So I'm planning on putting them in a box between the amp and
speakers with speakers in and out connectors on the back. Just
checking to see if anybody thinks this isn't worth doing for any
reason? From what little I know they should work fine like this,
it's just a little fiddly.

These are apparently 30A but that can be taken with a pinch of salt;
https://www.aliexpress.com/item/upc1237-large-current-speaker-protection-board/32809522760.html
(I purchased what look like identical items from a different seller
who no longer has them for sale. I see that listing says maximum amp
20w - that seems as unlikey as the previous listings 600w.)

Well I finally got a transformer (EI as I didn't think a toroidal was needed
as the 'signal' is high-current) and built two of these single channel
protection PCBs into a re-puposed case. I figurd they're better than nothing
(if only for the built-in 3 second delay to avoid 'thump') until I can
rebuild a different two-channel upc1237 based PCB with remote relays to
short the speakers...

The damn thing introduces shitloads of hum! It took me a while to make back
and front panels for the case so only just got around to hooking it up with
my existing amps. (I prioritised this over finishing the refurb of the other
amp so I'd have /some/ protection ready and could use it when finished.) I
tried removing the transformer and running it oiutside the case but it still
hums.

Biugger! I don't get much low-pain 'mobile' time as it is after the
housework etc is done and I've just wasted a bunch building this thing.
Who'd have thought these would introduce so much hum? (Not me obviously.)
It's not as if I've got my stereo system laid out as I'd like it yet with
speaker wires and power cables separated AMAP and only crossing at
right-angles (I'm planning on putting it in a different place when it's all
done so...).


**Ensure that the mains earth is isolated from the speaker earth/s.

--
Trevor Wilson
www.rageaudio.com.au

Phil Allison
Guest

Sat Jul 15, 2017 9:01 am   



Trevor Wilson wrote:

----------------------------

Quote:

**Ensure that the mains earth is isolated from the speaker earth/s.


** Also ensure the speaker grounds are kept separate too.

Wanna bet "misfit" has used a metal panel with 1/4 inch jacks for the speakers ?

In any case, he has created a totally unnecessary ground loop.



..... Phil

Trevor Wilson
Guest

Sat Jul 15, 2017 12:19 pm   



On 15/07/2017 5:01 PM, Phil Allison wrote:
Quote:
Trevor Wilson wrote:

----------------------------


**Ensure that the mains earth is isolated from the speaker earth/s.


** Also ensure the speaker grounds are kept separate too.

Wanna bet "misfit" has used a metal panel with 1/4 inch jacks for the speakers ?

In any case, he has created a totally unnecessary ground loop.


**I ain't taking that bet. Such a fault could be very nasty if the amp
is a bridged one.


--
Trevor Wilson
www.rageaudio.com.au

~misfit~
Guest

Sat Jul 15, 2017 12:27 pm   



Once upon a time on usenet Trevor Wilson wrote:
Quote:
On 15/07/2017 2:15 PM, ~misfit~ wrote:
Once upon a time on usenet ~misfit~ wrote:
[snipped]
The issue I have is that it has no speaker protection and my
speakers are irreplaceable (rarity and my budget considered). I
have two speaker protection PCBs sourced from AliExpress that
look up to the task, well designed with heavy duty relays and
thick signal traces made thicker by having the solder mask
removed and solder added. My problem is that I can't fit them and
their power supply into the amplifier case.

So I'm planning on putting them in a box between the amp and
speakers with speakers in and out connectors on the back. Just
checking to see if anybody thinks this isn't worth doing for any
reason? From what little I know they should work fine like this,
it's just a little fiddly.

These are apparently 30A but that can be taken with a pinch of salt;
https://www.aliexpress.com/item/upc1237-large-current-speaker-protection-board/32809522760.html
(I purchased what look like identical items from a different seller
who no longer has them for sale. I see that listing says maximum amp
20w - that seems as unlikey as the previous listings 600w.)

Well I finally got a transformer (EI as I didn't think a toroidal
was needed as the 'signal' is high-current) and built two of these
single channel protection PCBs into a re-puposed case. I figurd
they're better than nothing (if only for the built-in 3 second delay
to avoid 'thump') until I can rebuild a different two-channel
upc1237 based PCB with remote relays to short the speakers...

The damn thing introduces shitloads of hum! It took me a while to
make back and front panels for the case so only just got around to
hooking it up with my existing amps. (I prioritised this over
finishing the refurb of the other amp so I'd have /some/ protection
ready and could use it when finished.) I tried removing the
transformer and running it oiutside the case but it still hums.

Biugger! I don't get much low-pain 'mobile' time as it is after the
housework etc is done and I've just wasted a bunch building this
thing. Who'd have thought these would introduce so much hum? (Not me
obviously.) It's not as if I've got my stereo system laid out as I'd
like it yet with speaker wires and power cables separated AMAP and
only crossing at right-angles (I'm planning on putting it in a
different place when it's all done so...).


**Ensure that the mains earth is isolated from the speaker earth/s.


Thanks. It is. In both the amplifiers and the speaker protection unit.
(Although the amp I'm rewiring curently to use soon, the Playmaster Pro III
doesn't isolate the mains and speaker earths.)
--
Shaun.

"Humans will have advanced a long, long way when religious belief has a cozy
little classification in the DSM*."
David Melville (in r.a.s.f1)
(*Diagnostic and Statistical Manual of Mental Disorders)

~misfit~
Guest

Sat Jul 15, 2017 12:32 pm   



Once upon a time on usenet Phil Allison wrote:
Quote:
Trevor Wilson wrote:

----------------------------


**Ensure that the mains earth is isolated from the speaker earth/s.


** Also ensure the speaker grounds are kept separate too.

Wanna bet "misfit" has used a metal panel with 1/4 inch jacks for the
speakers ?

In any case, he has created a totally unnecessary ground loop.


I'll take the bet.

The back panel is hardboard and I have fixed 40cm wires going out the back
for 'speakers in' terminated in 3mm banana plugs. There are speaker jacks
mounted on the back panel that accept banana plugs, lugs or bare wire to the
speakers (currently using banana plugs).

Thanks.
--
Shaun.

"Humans will have advanced a long, long way when religious belief has a cozy
little classification in the DSM*."
David Melville (in r.a.s.f1)
(*Diagnostic and Statistical Manual of Mental Disorders)

Phil Allison
Guest

Sun Jul 16, 2017 12:03 pm   



~misfit~ wrote:

-------------------


Quote:

**Ensure that the mains earth is isolated from the speaker earth/s.

Thanks. It is. In both the amplifiers and the speaker protection unit.
(Although the amp I'm rewiring curently to use soon, the Playmaster Pro III
doesn't isolate the mains and speaker earths.)


** That is not what TW meant.

He was saying not to create and EXTRA mains earth going to the speaker ground terminals - cos doing that can make a "hum loop".



...... Phil

Phil Allison
Guest

Sun Jul 16, 2017 12:08 pm   



~misfit~ wrote:

--------------------


Quote:

**Ensure that the mains earth is isolated from the speaker earth/s.


** Also ensure the speaker grounds are kept separate too.

Wanna bet "misfit" has used a metal panel with 1/4 inch jacks for the
speakers ?

In any case, he has created a totally unnecessary ground loop.

I'll take the bet.

The back panel is hardboard and I have fixed 40cm wires going out the back
for 'speakers in' terminated in 3mm banana plugs. There are speaker jacks
mounted on the back panel that accept banana plugs, lugs or bare wire to the
speakers (currently using banana plugs).



** Have you somehow linked the speaker grounds from each channel ?

Cos if you have kept the channels separate and insulated from ground, the new speaker wiring cannot possibly cause the system to hum.



...... Phil

Phil Allison
Guest

Sun Jul 16, 2017 12:39 pm   



Trevor Wilson wrote:

---------------------------

Quote:

In any case, he has created a totally unnecessary ground loop.


**I ain't taking that bet. Such a fault could be very nasty if the amp
is a bridged one.



** Did I ever tell you the story of the "Hertz" Mosfet amplifiers and the Australian Navy ?

Hertz amplifiers are ( were ?) large, heavy, lateral TO3 mosfet beasts intended for the live music and disco markets. The ones I saw did 800W comfortably into 4 ohms at any phase angle you liked.

The local agent for the product sold a couple to the RAN, for use with SONAR testing in the submarine labs here in Sydney. He had blithely assumed that this was a job the amps could do easily.

But he had not reckoned on the pure ham fistedness of the RAN technicians.

Live Rock music and disco material can be pretty hard on an audio amplifier, but not near as lethal as a half trained Navy tech with a signal generator that goes to 100kHz - PLUS no clue at all about how audio gear is normally treated and not treated.

The Hertz amps has response up to 100kHz but NO WAY was that allowable at full power !!! Even 20kHz is pushing it due to the limited power handling of the output ZOBEL.

PLUS, the Navy decided to build a "path panel" for all inputs and outputs - a flat aluminium plate covered in 1/4 inch jack sockets !!!!

The idea of using a sheet of insulation material never crossed their tiny minds OR that 1/4 inch jacks and plugs were not up to handling 800W safely - OR that jack plug tips connect signal to the ground circuit FIRST every time they are plugged in.

The Hertz amps did not last long, soon as the tip of a jack carrying the output from the amp touched the patch panel, the amp blew up. Full output current ( about 50 amps) passed along the thin wires and tiny PCB tracks from the input XLR to common ground and vaporised.

The other amp in the pair soon had burnt film caps in the Zobel network.

After some tedious repairs, I wound up adding 2.2 ohm resistors in series with the input ground and bypassing XLR pin 1 to chassis with a 35amp bridge rectifier wired as back to back diodes. I did both amps while I hade them in the workshop.

They both passed the "Navy test" after that.

Not much I could do about some idiot feeding 5Vrms from an audio gen into the input at 100kHz though.

Another example was loaned to some lunatic in the NT to do "magnetics " experiments with - I think he was trying to make a linear motor. He succeeded in blowing the output MOSFETs to bits.

The Navy later succeeded in doing much the same to one of theirs.

Hertz amps had no DC rail fuses, juts thermal breakers and zener diode current limiting on the MOSFETs.

No way, whatsoever were they SAFE to use in a SONAR test lab.



..... Phil

Trevor Wilson
Guest

Sun Jul 16, 2017 10:35 pm   



On 16/07/2017 8:39 PM, Phil Allison wrote:
Quote:
Trevor Wilson wrote:

---------------------------


In any case, he has created a totally unnecessary ground loop.


**I ain't taking that bet. Such a fault could be very nasty if the amp
is a bridged one.



** Did I ever tell you the story of the "Hertz" Mosfet amplifiers and the Australian Navy ?

Hertz amplifiers are ( were ?) large, heavy, lateral TO3 mosfet beasts intended for the live music and disco markets. The ones I saw did 800W comfortably into 4 ohms at any phase angle you liked.

The local agent for the product sold a couple to the RAN, for use with SONAR testing in the submarine labs here in Sydney. He had blithely assumed that this was a job the amps could do easily.

But he had not reckoned on the pure ham fistedness of the RAN technicians.

Live Rock music and disco material can be pretty hard on an audio amplifier, but not near as lethal as a half trained Navy tech with a signal generator that goes to 100kHz - PLUS no clue at all about how audio gear is normally treated and not treated.

The Hertz amps has response up to 100kHz but NO WAY was that allowable at full power !!! Even 20kHz is pushing it due to the limited power handling of the output ZOBEL.

PLUS, the Navy decided to build a "path panel" for all inputs and outputs - a flat aluminium plate covered in 1/4 inch jack sockets !!!!

The idea of using a sheet of insulation material never crossed their tiny minds OR that 1/4 inch jacks and plugs were not up to handling 800W safely - OR that jack plug tips connect signal to the ground circuit FIRST every time they are plugged in.

The Hertz amps did not last long, soon as the tip of a jack carrying the output from the amp touched the patch panel, the amp blew up. Full output current ( about 50 amps) passed along the thin wires and tiny PCB tracks from the input XLR to common ground and vaporised.

The other amp in the pair soon had burnt film caps in the Zobel network.

After some tedious repairs, I wound up adding 2.2 ohm resistors in series with the input ground and bypassing XLR pin 1 to chassis with a 35amp bridge rectifier wired as back to back diodes. I did both amps while I hade them in the workshop.

They both passed the "Navy test" after that.

Not much I could do about some idiot feeding 5Vrms from an audio gen into the input at 100kHz though.

Another example was loaned to some lunatic in the NT to do "magnetics " experiments with - I think he was trying to make a linear motor. He succeeded in blowing the output MOSFETs to bits.

The Navy later succeeded in doing much the same to one of theirs.

Hertz amps had no DC rail fuses, juts thermal breakers and zener diode current limiting on the MOSFETs.

No way, whatsoever were they SAFE to use in a SONAR test lab.


**YIKES!

Did you ever see any of the big Mike Davis designed EHT MOSFET amps?
Best MOSFET amps I ever saw. Very clever design. Tough, reliable and
half decent sounding. MUCH better than the Perreaux crap back then (ca.
1980-something). He gave me some PCBs and I retro-fitted them to a
number of very troublesome amps. Never saw them back in the workshop.
The worst was this disgusting Musical Fidelity P370. Claimed to be 185
Watts Class A (actually, 10 Watts Class A). A shit-load of flat pack
MOSFETs and a horrible design. When one channel failed, both channels
went down, emitting much smoke in the process. EVERY SINGLE MOSFET
failed. I re-built it with a couple of Mike's modules. The customer
reported that it had never sounded so good. Reliable too.


--
Trevor Wilson
www.rageaudio.com.au

~misfit~
Guest

Mon Jul 17, 2017 2:09 am   



Once upon a time on usenet Phil Allison wrote:
Quote:
~misfit~ wrote:

--------------------



**Ensure that the mains earth is isolated from the speaker earth/s.


** Also ensure the speaker grounds are kept separate too.

Wanna bet "misfit" has used a metal panel with 1/4 inch jacks for
the speakers ?

In any case, he has created a totally unnecessary ground loop.

I'll take the bet.

The back panel is hardboard and I have fixed 40cm wires going out
the back for 'speakers in' terminated in 3mm banana plugs. There are
speaker jacks mounted on the back panel that accept banana plugs,
lugs or bare wire to the speakers (currently using banana plugs).



** Have you somehow linked the speaker grounds from each channel ?

Cos if you have kept the channels separate and insulated from ground,
the new speaker wiring cannot possibly cause the system to hum.


That's what I thought and why I was careful about grounds (I used to do
set-up, soundmixing and stage lighting for a band 'on the road' for a few
years so am aware of ground loops).

Yet still they introduce hum. I was wondering if perhaps my power
transformer is a little on the small side (as they are large relays) and if
so if that might be the cause. I have no experience using these
UPC1237-based protection circuits.

Cheers,
--
Shaun.

"Humans will have advanced a long, long way when religious belief has a cozy
little classification in the DSM*."
David Melville (in r.a.s.f1)
(*Diagnostic and Statistical Manual of Mental Disorders)

Phil Allison
Guest

Mon Jul 17, 2017 2:20 am   



Trevor Wilson wrote:
---------------------

Quote:


No way, whatsoever were they SAFE to use in a SONAR test lab.


**YIKES!

Did you ever see any of the big Mike Davis designed EHT MOSFET amps?



** Yep - saw quite a few of them, EHT2000 and EHT4000 models.

I managed to get a loan of an early production example of a 2000 from the owner who worked for Etone at the time. The chief designer for Jands, Doug Ford and his production manager Terry popped over to see it too.

The use of standard aluminium extrusions and fan blowing the toroidals transformers was commented on favourably. Having reed relay current sensors in the speaker lines was novel as well.

The way the AC wiring was installed on top of the heatsinks using a PCB and large plastic bolts was slightly mind blowing. But the amps had a fatal flaw that turned up after a few years in use.

The fan blew directly onto the copper pattern of the power amps PCBs which were *NOT* coated to protect them from moisture and the corrosion that results. Also, the lack of thermal cut offs meant that if the heatsinks clogged with fluff or the fan stopped, the whole assembly cooked.

Dissasembly, full immersion cleaning and repair is a MASSIVE job that I have done a few times, but not happily.

Perreaux had a similar problem with lack of PCB coating - but was far easier to deal with. Failure to deal with it meant the amps blew up as "tracking" began between adjacent tracks that differed by 200VDC or so. At least Perreauxs all had DC rail fuses that saved the MOSFETS.


..... Phil

~misfit~
Guest

Mon Jul 17, 2017 2:20 am   



Once upon a time on usenet Phil Allison wrote:
Quote:
~misfit~ wrote:

-------------------



**Ensure that the mains earth is isolated from the speaker earth/s.

Thanks. It is. In both the amplifiers and the speaker protection
unit. (Although the amp I'm rewiring curently to use soon, the
Playmaster Pro III doesn't isolate the mains and speaker earths.)


** That is not what TW meant.

He was saying not to create and EXTRA mains earth going to the
speaker ground terminals - cos doing that can make a "hum loop".


Cheers,
--
Shaun.

"Humans will have advanced a long, long way when religious belief has a cozy
little classification in the DSM*."
David Melville (in r.a.s.f1)
(*Diagnostic and Statistical Manual of Mental Disorders)

Goto page Previous  1, 2, 3, 4  Next

elektroda.net NewsGroups Forum Index - Electronics AUS - Speaker Protection Question.

Ask a question - edaboard.com

Arabic version Bulgarian version Catalan version Czech version Danish version German version Greek version English version Spanish version Finnish version French version Hindi version Croatian version Indonesian version Italian version Hebrew version Japanese version Korean version Lithuanian version Latvian version Dutch version Norwegian version Polish version Portuguese version Romanian version Russian version Slovak version Slovenian version Serbian version Swedish version Tagalog version Ukrainian version Vietnamese version Chinese version Turkish version
EDAboard.com map