Audio Relays

[jrmymllr]

I have this 18 year old Sony 5.1 receiver that's had a problem with one speaker for a while. When the power is first turned on, one speaker is very quiet. We've learned if the volume is turned up, then back down, it makes it work until next time.

I have the service manual for this thing and recently I decided to tear into it and fix it once and for all. I expected it to be some capacitor or semiconductor thing but no, it's the speaker output relay...the relay right before the rear speaker jacks. Cool, that's easy to fix. Apparently this is a somewhat common symptom and temporary fix I've been doing.

The original relay is a ALA2F12 and the datasheet is here: https://datasheet.octopart.com/ALA2F12-Panasonic-datasheet-11734546.pdf

The part is obsolete but Digikey claims OSA-SS-212DM3,000 (https://www.te.com/usa-en/product-5-1419124-6.datasheet.pdf) is the replacement. Curious, I compared the two datasheets. I noticed the original relay is "best suited for switching speakers".......I was then reminded all relays are not the same if one ignores the voltage and current ratings. The original contacts are gold clad, AgNi contacts, and the replacement is "Ag Alloy". I also found it curious that the original relays appear to be open to the air, and I'm not so sure about the replacement.

1. What do you think about the replacement relay? Is this type of contact material not suited for this application for long term reliability?
2. Should I pre-emptively replace the other 3 identical relays while I'm at it? In other words, is it likely they are well on the road to failure given the same part number and same application?

[andy3055]

What makes you think it is the rely? Make sure there is no dry solder joint on any of the pins first. If that does not help, interchange with the relay from the "good" channel. If the fault shifts with the so called "bad" relay, you can say it is bad. If not, you will have to check other possibilities.

[ambrosia heart]

No time to read your long composition. Your problem can be fixed as below stated.

It is normal for the relay 4 contact points to have oxides after 5-6 use.

Steps:
No need to replace it.
Just open the circled 2 latches 
You will find 4 contacts points 
There are some oxides on them. 
Use a sand paper to remove the oxide and test the
conductivity of 4 points by a digital meter (zero ohm)
 when press the sliver plane. 
cover the relay again

see pm to see inside structure of the relay.

The long essay and photos was written and taken by me.

[floobydust]

Sometimes (due to humidity) you can get a passivation layer on relay contacts, switches, connectors etc. which sounds terrible- gritty, distorted, volume up and down, poor connection that (for loudpseaker connections) mysteriously vanishes when you crank it up.

Clean relay contacts using a piece of paper for the abrasive. Just pull an inch of paper through contacts that are gently pressed closed. Sandpaper is too harsh and removes any plating. For low-level signal relays I also put PPE contact treatment on the contacts afterwards.

There is a chance it's not the relay but a problem with a semiconductor or electrolytic capacitor in the power amp.

[bob91343]

The fact that turning up the volume fixes the problem, points to oxidation of relay contacts.  Cleaning will resolve the problem for a long time.  When the contacts close, there may be little or no signal present and thus nothing to break through the corrosion.  Sometimes increasing the contact pressure helps, perhaps by a higher coil voltage.  At least at first.  Prove it by tapping the relay next time it fails.

[wraper]

No time to read your long composition. Your problem can be fixed as below stated.

It is normal for the relay 4 contact points to have oxides after 5-6 use.

Steps:
No need to replace it.
Just open the circled 2 latches 
You will find 4 contacts points 
There are some oxides on them. 
Use a sand paper to remove the oxide and test the
conductivity of 4 points by a digital meter (zero ohm)
 when press the sliver plane. 
cover the relay again

see pm to see inside structure of the relay.

The long essay and photos was written and taken by me.

And the result is unreliable trash.

There are some oxides on them.

More likely that contacts are damaged by arcing and you remove contact material to make it less bad.

[Whales]

I've encountered exactly this before.  Turning the volume up & playing loud sounds or banging the unit with my hand would fix it.  Eventually neither was enough and I cracked open the relay to clean the contacts.  IIRC they looked blackened.

Relay & switch contacts are typically only rated to make & break within a certain current range (eg 100mA to 1A) called a "wetting current".  Anything outside that range will dirty or erode the contacts.  Different contact materials (silver, gold, alloys, etc) are used to vary the wetting current range. 

Often amplifiers use relays to hide the turn-on-thump noise.  Unfortunately this means that the speaker side of the relay contacts turns on & off when there is very little current flowing, ie probably outside of wetting limits.  Once oxide builds up: they probably start running hot, which might explain why mine were blackened?

[Gyro]

Speaker relay contacts should never look blackened from electrical damage - the power levels simply aren't there (unless it's a big PA amp and stack), and they should be closing with zero voltage / current flowing (or you'd get a very loud pop). That's the problem, no wetting current when closing.

The most likely reason for blackening is oxidation of pure silver or silver alloy contacts. This is much more likely if there is any Sulphur content in the air, but will eventually happen, regardless, unless the atmospheric humidity is zero.

Use a sand paper to remove the oxide and test the
conductivity of 4 points by a digital meter (zero ohm)

As mentioned, using sandpaper on relay contacts is terrible advice, which hasn't be relevant for decades! It has no place other than maybe old car ignition solid tungsten 'points'. Certainly nowhere that plated or flashed relays are used. Even on solid contact material, that sort of contact abrasion greatly reduces the area of the closing contacts, seriously reducing their current capacity and encouraging welding.

As floobydust said, you should never use anything more abrasive than ordinary paper, preferably soaked in IPA or contact cleaner.

[Le_Bassiste]

Speaker relay contacts should never look blackened from electrical damage - the power levels simply aren't there (unless it's a big PA amp and stack), and they should be closing with zero voltage / current flowing (or you'd get a very loud pop). That's the problem, no wetting current when closing.

The most likely reason for blackening is oxidation of pure silver or silver alloy contacts. This is much more likely if there is any Sulphur content in the air, but will eventually happen, regardless, unless the atmospheric humidity is zero.

As mentioned, using sandpaper on relay contacts is terrible advice, which hasn't be relevant for decades! It has no place other than maybe old car ignition solid tungsten 'points'. Certainly nowhere that plated or flashed relays are used. Even on solid contact material, that sort of contact abrasion greatly reduces the area of the closing contacts, seriously reducing their current capacity and encouraging welding.

As floobydust said, you should never use anything more abrasive than ordinary paper, preferably soaked in IPA or contact cleaner.

^this.

[jrmymllr]

What makes you think it is the rely? Make sure there is no dry solder joint on any of the pins first. If that does not help, interchange with the relay from the "good" channel. If the fault shifts with the so called "bad" relay, you can say it is bad. If not, you will have to check other possibilities.

I know it's the relay because I've already reflowed the joints and measured resistance across the contacts when the relay is pulled in. One side is good, the other floats around 5-100 ohms.

[jrmymllr]

As mentioned, using sandpaper on relay contacts is terrible advice, which hasn't be relevant for decades! It has no place other than maybe old car ignition solid tungsten 'points'. Certainly nowhere that plated or flashed relays are used. Even on solid contact material, that sort of contact abrasion greatly reduces the area of the closing contacts, seriously reducing their current capacity and encouraging welding.

As floobydust said, you should never use anything more abrasive than ordinary paper, preferably soaked in IPA or contact cleaner.

Is cleaning these contacts a long term solution? Or is it likely to need it again in a year? For the price I don't mind replacing because getting the PCB out is a bit of a job.

Or, crazy thought here, would it be unwise to apply an external voltage through a resistance which would limit to 0.5 or 1 amp, while I make the relay energize a few times to clean the contacts? Since this relay is literally at the very end of the amp output, I could apply an isolated external voltage without the worry that it'll flow through some other circuit in the receiver.

I only suggest this because it would be so easy to do.

[wraper]

Speaker relay contacts should never look blackened from electrical damage - the power levels simply aren't there (unless it's a big PA amp and stack), and they should be closing with zero voltage / current flowing (or you'd get a very loud pop). That's the problem, no wetting current when closing.

You will not get a pop, since there is only AC voltage when relay activates. For pop to happen you need DC.

[jrmymllr]

And the result is unreliable trash.

There are some oxides on them.

More likely that contacts are damaged by arcing and you remove contact material to make it less bad.

Here's another thought: the original relay contacts are gold plated. Gold does not tarnish, correct? So if this is occurring wouldn't that mean the gold plating is wore off and any repair to the relay would be short-lived?

[Gyro]

Speaker relay contacts should never look blackened from electrical damage - the power levels simply aren't there (unless it's a big PA amp and stack), and they should be closing with zero voltage / current flowing (or you'd get a very loud pop). That's the problem, no wetting current when closing.

You will not get a pop, since there is only AC voltage when relay activates. For pop to happen you need DC.

Hah, good point. Although even with AC the power levels on relay closing are still tiny at normal listening levels (around 1W typically). Personally, I never start a source before the amplifier anyway, otherwise you miss the beginning.

[Gyro]

And the result is unreliable trash.

There are some oxides on them.

More likely that contacts are damaged by arcing and you remove contact material to make it less bad.

Here's another thought: the original relay contacts are gold plated. Gold does not tarnish, correct? So if this is occurring wouldn't that mean the gold plating is wore off and any repair to the relay would be short-lived?

It depends, sometimes the plating is so thin that you get micro-fractures and the underlying material typically Silver 'growing' though. Yes mechanical wear can also rub through the plating.

Cleaning the contacts could well give a decent life extension, but there's no doubt that using higher quality, preferably sealed, relays would give a longer trouble-free life.

[ambrosia heart]

And the result is unreliable trash.

There are some oxides on them.

More likely that contacts are damaged by arcing and you remove contact material to make it less bad.

Here's another thought: the original relay contacts are gold plated. Gold does not tarnish, correct? So if this is occurring wouldn't that mean the gold plating is wore off and any repair to the relay would be short-lived?

No time to see other. Only answer you!!
Gold plating??? Are you insane??!! only 12HKD. (see photo)
Why black in color?
See chemical reactions (photo)

Is cleaning these contacts a long term solution?
Of course , Not.  Even you buy a new one, same
consquence happens after 5-6 year.  Why due to N2
 and spark. You can avoid it by making relay fully filled with
inert as, Neon only.

Or is it likely to need it again in a year? For the price I don't mind replacing because getting the PCB out is a bit of a job.
After 5-6 year, same consequence occurs due to
chemical reactions (see photo).  Just try it by removing
the Cu(II)O from 4 contacts by a sand paper. Your amp
will work well within 5 years.

[jrmymllr]

Here's another thought: the original relay contacts are gold plated. Gold does not tarnish, correct? So if this is occurring wouldn't that mean the gold plating is wore off and any repair to the relay would be short-lived?

No time to see other. Only answer you!!
Gold plating??? Are you insane??!! only 12HKD. (see photo)
Why black in color?
See chemical reactions (photo)

Is cleaning these contacts a long term solution?
Of course , Not.  Even you buy a new one, same
consquence happens after 5-6 year.  Why due to N2
 and spark. You can avoid it by making relay fully filled with
inert as, Neon only.

Or is it likely to need it again in a year? For the price I don't mind replacing because getting the PCB out is a bit of a job.
After 5-6 year, same consequence occurs due to
chemical reactions (see photo).  Just try it by removing
the Cu(II)O from 4 contacts by a sand paper. Your amp
will work well within 5 years.

Not sure what you mean about the gold plating, but from what I can tell the original ALA2F12 relays are gold clad. See screenshot.

[Smith]

Just replaced on of these last weekend. I mostly see them fail on the rear channels. If you crank them up they will work again, but they will soon fail again. Sometimes bad relays even make the amp sound distorted. Relays do seem to fail al lot after 10-20 years, especially when you don't drive it that hard. Even on expensive stereo amps, I replaced a lot of them. I try to use good brand relays, especially with gold plated contacts. Most of the time they aren't that much more expensive. If you want to know if it's the relay, one way is to shutdown and startup the amp (or disable and enable the speakers, or mute them if your amp switches the relays on these settings) , and listen if the sound comes back when the relays click on again. If the sound does come back, even for a very short while the relay is probable shot.

On the last repair I tried sending some current trough them, about 1-2amps DC. I wanted to try this because the amp is old and cheap and wasn't worth trowing any money at. After this "repair" the relays measured perfect, but 3 months later the same relay failed again. Luckily I still had some of the same relays left from another amp. Just replace it, only "fix" it with sandpaper if you can't wait for new ones to arrive.

[jrmymllr]

Just replaced on of these last weekend. I mostly see them fail on the rear channels. If you crank them up they will work again, but they will soon fail again. Sometimes bad relays even make the amp sound distorted. Relays do seem to fail al lot after 10-20 years, especially when you don't drive it that hard. Even on expensive stereo amps, I replaced a lot of them. I try to use good brand relays, especially with gold plated contacts. Most of the time they aren't that much more expensive. If you want to know if it's the relay, one way is to shutdown and startup the amp (or disable and enable the speakers, or mute them if your amp switches the relays on these settings) , and listen if the sound comes back when the relays click on again. If the sound does come back, even for a very short while the relay is probable shot.

On the last repair I tried sending some current trough them, about 1-2amps DC. I wanted to try this because the amp is old and cheap and wasn't worth trowing any money at. After this "repair" the relays measured perfect, but 3 months later the same relay failed again. Luckily I still had some of the same relays left from another amp. Just replace it, only "fix" it with sandpaper if you can't wait for new ones to arrive.

Good timing. I just tried putting 1A through this one. I cycled the relay several times with 12V applied through 10 ohms. The 'on' resistance improved drastically to a fraction of an ohm, then a few minutes later was back up to over 2 ohms. Blahhgg. So I will be replacing them.

I wonder if a sealed silver alloy contact relay (like the TE part I quote above) would be ok. Maybe if I turn up the volume occasionally and make that relay cycle to burn off whatever forms on there!

EDIT: just looked through Digikey and while I'd want a "signal relay", they are rated for less than the 3A the original is. So that leaves "power relays" and none have gold clad contacts.

A 2A signal relay with the correct basic parameters has the wrong footprint. Might have to do the silver contact route.

[Gyro]

Or is it likely to need it again in a year? For the price I don't mind replacing because getting the PCB out is a bit of a job.
After 5-6 year, same consequence occurs due to
chemical reactions (see photo).  Just try it by removing
the Cu(II)O from 4 contacts by a sand paper. Your amp
will work well within 5 years.

Just for reference, no relays use bare copper contacts these days. Attacking the contacts with sandpaper is just that!

[Gyro]

I wonder if a sealed silver alloy contact relay (like the TE part I quote above) would be ok. Maybe if I turn up the volume occasionally and make that relay cycle to burn off whatever forms on there!

EDIT: just looked through Digikey and while I'd want a "signal relay", they are rated for less than the 3A the original is. So that leaves "power relays" and none have gold clad contacts.

A 2A signal relay with the correct basic parameters has the wrong footprint. Might have to do the silver contact route.

Silver contacts can be very good (lowest resistance), but do need a wiping action to keep them clean at low currents. Some relays achieve a degree of flex and wiping action. Sealed silver contact relays are better but I wouldn't guarantee that they will be reliable for signal switching (unless they are spec'd to do it).

You may have to be prepared to abandon the footprint and use wires to connect from the PCB footprint to the relay to achieve the most reliable result.


EDIT: Something like the Panasonic S series relay (available from Farnell among others) might be a useful choice. Gold clad Silver with good (specified) low current performance. Ideally get one with 4 contacts and use parallel pairs. Either that or use 2 relays and parallel all of the contacts...  https://uk.farnell.com/w/c/switches-relays/relays/power-relays?product-range=s-series

[Smith]

Just for reference, no relays use bare copper contacts these days. Attacking the contacts with sandpaper is just that!

Indeed. The copper will deteriorate at a much faster rate than before. It will work, but the problem will probably come back just as bad or even worse.

Silver contacts can be very good (lowest resistance), but do need a wiping action to keep them clean at low currents. Some relays achieve a degree of flex and wiping action. Sealed silver contact relays are better but I wouldn't guarantee that they will be reliable for signal switching (unless they are spec'd to do it).

Silver is indeed much better than bare copper, but silver will corrode too. The best option is gold. There is a reason the best connectors for small signals are gold plated. This is a great conductor, specially for small signals and it will not corrode nearly as fast as silver of even bare copper will. Special relays for very low signals have gold contacts too for a reason.

[shakalnokturn]

Having been through a fair deal of audio amp repairs that at least partly involved old output relays, I'd agree to say that defective ones are almost always the non-sealed silver contact ones.
Cleaning them with contact cleaner impregnated paper will usually restore them for years.
If you want longer life than the original go for gold or sealed.

[jrmymllr]

Having been through a fair deal of audio amp repairs that at least partly involved old output relays, I'd agree to say that defective ones are almost the non-sealed silver contact ones.
Cleaning them with contact cleaner impregnated paper will usually restore them for years.
If you want longer life than the original go for gold or sealed.

So sealed silver alloy would do the trick you think? Also since this is already 18 years old and was not expensive when new, I suppose  another 5 years would be doing well.

[Gyro]

I'm not sure if you saw my edit about the Panasonic S series relays on my previous post - they will cost you though, so if the amp wasn't that expensive to start with, they're probably not an economical option.