the relay you put in place to protect your speakers from amp failure might fail itself, thus taking out the speakers
dc protection with relays
Guess what?
Shocker of shockers, any device you use can fail, Mosfets included. I have seen relays fail to open, but the energies involved were extreme and the relay not selected properly. So let's not point at rare occurrences. You are obviously invested in supporting a Mosfet solution. What I am interested in is recommending the most appropriate solution for this situation.
Show me a great Mosfet based solution and I may use it in new builds or problem situations as long as the price is within reason. Otherwise, relays are exactly the right device to use. It's also what was originally installed, lasted decades and an easy repair.
I've seen far more speakers turn into smoking ruins and little fires with the use of fuses (terrible idea) or nothing at all. Speaker failure due to relay failure is rare when the relay is the correct one. The only real issue with relays are contact problems because the operator allows it to open or close while passing energy. Now, one brand was more common for welding speaker relay contacts. The larger Marantz receivers. They could really dump current! But to make a Marantz fail .... well, you deserve it and your speakers are probably just hanging on by a thread anyway! lol!
Been in this business for nearly 50 years. I've seen quite a lot. Being in the Toronto area and in pro sound also, I've seen more than you probably realize.
Shocker of shockers, any device you use can fail, Mosfets included. I have seen relays fail to open, but the energies involved were extreme and the relay not selected properly. So let's not point at rare occurrences. You are obviously invested in supporting a Mosfet solution. What I am interested in is recommending the most appropriate solution for this situation.
Show me a great Mosfet based solution and I may use it in new builds or problem situations as long as the price is within reason. Otherwise, relays are exactly the right device to use. It's also what was originally installed, lasted decades and an easy repair.
I've seen far more speakers turn into smoking ruins and little fires with the use of fuses (terrible idea) or nothing at all. Speaker failure due to relay failure is rare when the relay is the correct one. The only real issue with relays are contact problems because the operator allows it to open or close while passing energy. Now, one brand was more common for welding speaker relay contacts. The larger Marantz receivers. They could really dump current! But to make a Marantz fail .... well, you deserve it and your speakers are probably just hanging on by a thread anyway! lol!
Been in this business for nearly 50 years. I've seen quite a lot. Being in the Toronto area and in pro sound also, I've seen more than you probably realize.
I should add, the article you linked to isn't really dealing with facts. For example, it doesn't matter if your speaker is wired to ground directly or through a relay. It is a series circuit, the contacts will burn if you cause an arc. The contacts don't care where they are in line.
Speakers will burn with 12 VDC or less across them.
Now, try mercury wetted relays. Those are really tough, they will interrupt the current and they are ghastly expensive. I use them in my dummy load speaker load switcher. Big ones, really expensive. I would have been further ahead with normal relays replacing them every few years "just to be sure".
Speakers will burn with 12 VDC or less across them.
Now, try mercury wetted relays. Those are really tough, they will interrupt the current and they are ghastly expensive. I use them in my dummy load speaker load switcher. Big ones, really expensive. I would have been further ahead with normal relays replacing them every few years "just to be sure".
Another thorny issue with relays is how distorting they are at high current (due to ferromagnetic parts carrying current). Not something you find on the datasheet, but for high performance equipment choosing the wrong relay might be the limiting part (also an issue with terminal posts or connectors that aren't brass). Of course your speakers are going to be distorting more, but if you want those low distortion figures on the specifications you may have to deal with this.
Hi Mark,
Yes, exactly right. As you mentioned this isn't an issue with the right relays. It all comes down to the right part for the right job.
So it really isn't a thorny issue at all. It's the same as using the wrong part anywhere. You're also right in that distortion in loudspeaker systems increases rapidly with power levels (excursion and voice coil temperature).
Yes, exactly right. As you mentioned this isn't an issue with the right relays. It all comes down to the right part for the right job.
So it really isn't a thorny issue at all. It's the same as using the wrong part anywhere. You're also right in that distortion in loudspeaker systems increases rapidly with power levels (excursion and voice coil temperature).
I think the problem people sometimes face in parts choice situation is they are aware of these basic criteria, but the myriad of available relay choices make it difficult to break it down to the appropriate type, make and model for a fault protection application. In this case, I’m referring to someone adding a protection stage to a new build, or a standalone unit for an amplifier having no output protection.
It would be useful to have a table showing relays that are presumed to be appropriate for given rail voltages and assumed load impedances.
Well then, you're asking for free knowledge on hard won experience. A lot depends on the application too (space, power, etc)
Put it simply. Larger current ratings sometimes come at a price. The coil will draw more current to overcome higher return spring tension. Either that or the contacts will not wipe and have as much closing force.
If you only need a 3 ampere relay, there is zero point in using one rated for 10 amperes. Most here probably would even though in that application the smaller relay may work better. With parts, everything is a trade-off. You balance attributes against each other for the best outcome and it isn't just price. The more expensive part is not always better either. Of course there is always size too. If it is a new build you can allocate room for it. Not so always with an existing item, home built or not.
Anyway, this thread is on an existing commercially available piece of equipment. We should limit our discussion to topics helpful to the OP.
Put it simply. Larger current ratings sometimes come at a price. The coil will draw more current to overcome higher return spring tension. Either that or the contacts will not wipe and have as much closing force.
If you only need a 3 ampere relay, there is zero point in using one rated for 10 amperes. Most here probably would even though in that application the smaller relay may work better. With parts, everything is a trade-off. You balance attributes against each other for the best outcome and it isn't just price. The more expensive part is not always better either. Of course there is always size too. If it is a new build you can allocate room for it. Not so always with an existing item, home built or not.
Anyway, this thread is on an existing commercially available piece of equipment. We should limit our discussion to topics helpful to the OP.
Just download the spec sheets from different makers for the requirement, and filter them for your needed application.
Make a list of three, find which one is in stock, and order those from your preferred supplier.
There are variations in coil voltage, housing size, open / covered / sealed / hermetically sealed, type of contact, current rating, and so on.
Making a table means a lot of work, at $50 and up an hour...
Make a list of three, find which one is in stock, and order those from your preferred supplier.
There are variations in coil voltage, housing size, open / covered / sealed / hermetically sealed, type of contact, current rating, and so on.
Making a table means a lot of work, at $50 and up an hour...
music, lets say 2 to 3 amps peak, your relay cant cope? no issue with mosfets.
when the amp fails, the current could be 10 amps continuous - how will your relay fare breaking that? no issue with mosfets.
output relays
It seems that the use of MOS-FET transistors is that they need a special PCB to be fabricated, and the parts sourced.
As with any modification, it may need to be modified / tweaked for optimal response, which I feel is above the OP's level of competence.
Most good relays will last 15 + years, so how old will you be then?
That is a lot of effort for a single $10 relay replacement.
As with any modification, it may need to be modified / tweaked for optimal response, which I feel is above the OP's level of competence.
Most good relays will last 15 + years, so how old will you be then?
That is a lot of effort for a single $10 relay replacement.
99.999% of the time the relays never have to break any current. Even with the most common amp failure mode - DC at the output at switch-on. This happens, the relay simply fails to close. In the event of a run time fault you will probably blow the mains fuse anyway. You may have to replace a fuse and output transistors. If the relay gets trashed just replace it, youre already spending money on transistors.
ctrlx,
You seem focused on a worst case that seldom happens. Relays often break currents of that magnitude in case you didn't know. Of course they get trashed when they do from an audio quality standpoint, but if they interrupted that amount of current, you probably have other issues as well. To inject reality, I have also seen shorted mosfets designed to disconnect a load. So what?
Honestly, you are so invested in using mosfets no matter what I don't know that you can objectively look at the issue.
You know where mosfets and triacs excel? Output crowbars. Yup, they can short, burn traces off and all kinds of fun.
You seem focused on a worst case that seldom happens. Relays often break currents of that magnitude in case you didn't know. Of course they get trashed when they do from an audio quality standpoint, but if they interrupted that amount of current, you probably have other issues as well. To inject reality, I have also seen shorted mosfets designed to disconnect a load. So what?
Honestly, you are so invested in using mosfets no matter what I don't know that you can objectively look at the issue.
You know where mosfets and triacs excel? Output crowbars. Yup, they can short, burn traces off and all kinds of fun.
I had a temperature controller system (4 in a set), each ran the contactor linked to it, about one cycle a minute, and that was through a miniature relay made by O/E/N locally.
They were taking the inputs from 4 zones through a sensor, and each was controlling the heater linked to it.
Two were replaced after fifteen years, and two were in service, 1993-2018, when I replaced the entire PLC, which had SSR outputs and thermo-couple inputs, making the controllers superfluous.
Work that out, how many operations.
Like Anatech says, you seem to have decided on a course of action, so go ahead, you will have only yourself to praise or blame for the result.
It is your property, our advice, based on many years' experience, has been given.
Your call.
They were taking the inputs from 4 zones through a sensor, and each was controlling the heater linked to it.
Two were replaced after fifteen years, and two were in service, 1993-2018, when I replaced the entire PLC, which had SSR outputs and thermo-couple inputs, making the controllers superfluous.
Work that out, how many operations.
Like Anatech says, you seem to have decided on a course of action, so go ahead, you will have only yourself to praise or blame for the result.
It is your property, our advice, based on many years' experience, has been given.
Your call.
Divide the amplifier's supply voltage by the loudspeaker's DC resistance and you know what DC current the relay has to be able to break.
Larger currents are possible when the output is shorted and the amplifier has no current limiter, but with shorted output, the loudspeaker doesn't need to be protected. You just need to keep the amplifier from going on fire, that's what fuses are for.
Larger currents are possible when the output is shorted and the amplifier has no current limiter, but with shorted output, the loudspeaker doesn't need to be protected. You just need to keep the amplifier from going on fire, that's what fuses are for.
Unfortunately there are a lot of threads to the same topic - probably because of a topic choice, that is too general like "best Audio relays" or "output relays" or "best relays for signal path" or "Power Relays" or "best-sounding-signal-level-relay" etc. (except this thread).
From my view there are a four kinds of relays in audio components and all of which needs its individual special features concerning contact material:
- relays for small signal applications (source select in Pre- and integrated amps so as for relay controlled attenuators for volume control),
- relays for very small signal requirements - e. g. for select MM- and MC- gain and load or for turntable select by use of only one RIAA stage,
- relays for switch on power transformers in remote controlled amplifiers so as inrush current limiters - and
- relays for the aim of voice coil resp. speaker protection against DC and hard popping while switch-on and switch-off.
For the last application (the highest demands are here, i. e. conflicting demands like excellent small signal and large signal behavior at the same time are required), a distinction is made between these versions:
solid state relays like
https://neurochrome.com/products/guardian-86
and usual mechanical relays.
One must browsing in detail all this threads, because just everywhere - i. e. in each thread are to find information from more than only one variant:
https://www.diyaudio.com/community/threads/output-relays.191449/
https://www.diyaudio.com/community/threads/loudspeaker-relays.120365/
https://www.diyaudio.com/community/threads/best-relays-for-signal-path.5906/
https://www.diyaudio.com/community/threads/recommended-relay-for-line-level-switch.328050/
https://www.diyaudio.com/community/threads/best-sounding-signal-level-relay.257868/
In post #11 under
https://www.diyaudio.com/community/threads/audio-switching-relay-recommendations.348883/
I have mentioned further threads from here concerning relays for all kind of applications.
I want to have a good and not too expensive universal speaker protect relay for good replacement in old amplifier stuff (top design solid state relays seems to be the royal way, but mostly too expensive).
If one want to deal with classic designs, i. e. with mechanical contacts, is a version with two contacts necessary, as already recommended 38 years ago in an Elektor magazine - this special edition with diy project: "Die Grönemeyer Box" (author: GJW = Guido J. Wasser) - go to
https://www.hifi-selbstbau.de/index.php?option=com_content&view=article&id=33
In the amplifier-part (German's Elektor "Sonderheft 4" from 1986) was recommend at those days a SDS relay for speaker protect with two different contact materials - one for large signal and one for small signal requirements.
Who know the exact naming of this no longer available relay from SDS and the successor type ?
Thank you very much for posting this information.
I have heard from some guys, that this Panasonic series would be a successor:
https://web.archive.org/web/2010082...gues/downloads/relays/ds_61005_0001_en_ds.pdf
https://na.industrial.panasonic.com...lineup/mechanical-signal-relays/series/119573
https://mediap.industry.panasonic.eu/assets/download-files/import/ds_ds_en_discon2012.pdf
But I'm unsure if that's true.
For new diy projects or more complex modifications on old amp stuff there is the possibility of using two or three independend relays (with its own driver transistor) with different contact materials and with different time delay for contact shorting - or as mentioned before, a top class solid state relay without audible distortion.
But I want for easy replacements in repair service a single relay compare to the old mention version from SDS.
The only current available relay with two different contact-material, which I know, is this one:
https://shop.dadaelectronics.eu/loudspeaker-relay-24v-lr-24-amplimo-108462111.html
https://www.hod-electronics.nl/de/relais/lautsprecher-relais/AMP LR-24V
also mentioned in the URL from post 21 here.
But I think, there are various other relay models in this kind especially since the fundamental issue on these relays for speaker protect has been known for many years.
From my view there are a four kinds of relays in audio components and all of which needs its individual special features concerning contact material:
- relays for small signal applications (source select in Pre- and integrated amps so as for relay controlled attenuators for volume control),
- relays for very small signal requirements - e. g. for select MM- and MC- gain and load or for turntable select by use of only one RIAA stage,
- relays for switch on power transformers in remote controlled amplifiers so as inrush current limiters - and
- relays for the aim of voice coil resp. speaker protection against DC and hard popping while switch-on and switch-off.
For the last application (the highest demands are here, i. e. conflicting demands like excellent small signal and large signal behavior at the same time are required), a distinction is made between these versions:
solid state relays like
https://neurochrome.com/products/guardian-86
and usual mechanical relays.
One must browsing in detail all this threads, because just everywhere - i. e. in each thread are to find information from more than only one variant:
https://www.diyaudio.com/community/threads/output-relays.191449/
https://www.diyaudio.com/community/threads/loudspeaker-relays.120365/
https://www.diyaudio.com/community/threads/best-relays-for-signal-path.5906/
https://www.diyaudio.com/community/threads/recommended-relay-for-line-level-switch.328050/
https://www.diyaudio.com/community/threads/best-sounding-signal-level-relay.257868/
In post #11 under
https://www.diyaudio.com/community/threads/audio-switching-relay-recommendations.348883/
I have mentioned further threads from here concerning relays for all kind of applications.
I want to have a good and not too expensive universal speaker protect relay for good replacement in old amplifier stuff (top design solid state relays seems to be the royal way, but mostly too expensive).
If one want to deal with classic designs, i. e. with mechanical contacts, is a version with two contacts necessary, as already recommended 38 years ago in an Elektor magazine - this special edition with diy project: "Die Grönemeyer Box" (author: GJW = Guido J. Wasser) - go to
https://www.hifi-selbstbau.de/index.php?option=com_content&view=article&id=33
In the amplifier-part (German's Elektor "Sonderheft 4" from 1986) was recommend at those days a SDS relay for speaker protect with two different contact materials - one for large signal and one for small signal requirements.
Who know the exact naming of this no longer available relay from SDS and the successor type ?
Thank you very much for posting this information.
I have heard from some guys, that this Panasonic series would be a successor:
https://web.archive.org/web/2010082...gues/downloads/relays/ds_61005_0001_en_ds.pdf
https://na.industrial.panasonic.com...lineup/mechanical-signal-relays/series/119573
https://mediap.industry.panasonic.eu/assets/download-files/import/ds_ds_en_discon2012.pdf
But I'm unsure if that's true.
For new diy projects or more complex modifications on old amp stuff there is the possibility of using two or three independend relays (with its own driver transistor) with different contact materials and with different time delay for contact shorting - or as mentioned before, a top class solid state relay without audible distortion.
But I want for easy replacements in repair service a single relay compare to the old mention version from SDS.
The only current available relay with two different contact-material, which I know, is this one:
https://shop.dadaelectronics.eu/loudspeaker-relay-24v-lr-24-amplimo-108462111.html
https://www.hod-electronics.nl/de/relais/lautsprecher-relais/AMP LR-24V
also mentioned in the URL from post 21 here.
But I think, there are various other relay models in this kind especially since the fundamental issue on these relays for speaker protect has been known for many years.
Last edited:
in post #1 unter
https://www.aktives-hoeren.de/viewtopic.php?t=3095
is to find the genuine part of the mentioned article, which you will find in the attached file.
The part of this article from post #1 here in English (some of what I wrote in my previous post from memory wasn't entirely correct in several details):
Guido J. Wasser claim : Relay Contacts: the weak point
The loudspeaker relays are more important than many electronics engineers think. This engineers usually claim:
"Thick contacts - if possible a car relay - for high currents, and gold plated to prevent oxidation."
Looks great but works only at the beginning in the right manner.
After a few weeks or months, the amplifier's damping factor goes down from 100 to 5 or worse.
The distortions have far exceeded especially at high frequencies (Figure 5) and low levels.
When switching under load for the first time, the tiny electric arc burned away the extremely thin layer of gold (check out fig. 2 first datasheet in my previous post).
If you continue to switch at currents of a few mA to amps, the contact will oxidize, but it will burn itself out again and again until it fails completely.
But very high currents does rarely flow; in most cases small currents are present.
Example: A drumbeat with full amplifier power reaches the peak value of 65 volts.
Around 11 amps then flow at the minimum impedance of 6 ohms.
A strong contact with low internal resistance is therefore required.
If internal contact resistance get high value of around 1 ohm, e.g. due to burnout, approximately 85 watts of power loss occurs at this contact;
This means, it becomes hot, resistance continues to rise, and oxidation continues to increase.
Contacts with internal resistances in the milliohm range with currents of around 10 amperes are now easy to create and also remain cold.
However, they only stay clean because the high current constantly burns them free.
At medium volumes of around 80 dB, however, only a few hundred milliamperes flow; In quiet areas the current drops to a few microamps.
Gold-plated contacts would be ideal for this, but unfortunately they burn away immediately at high currents.
The only thing that helps here is “Use both advantages without the disadvantages!”
I recommend a S relay from SDS. All four contacts are realized in multiple layers and I connect all four in parallel. Since one always pulls on first, only this contact burns freely under load. The other three contacts do not change and conduct medium and low currents.
During operation, when all switches are closed, the current is distributed across the four contacts and does not cause any harm.
My relays have already completed several thousand switching procedures with and without load, and subsequent measurements still show the same good data concerning the contact resistance as when they were in new condition.
I have heard this approach years ago from an American high-end manufacturer, and SDS S4 relays in the meantime available from SDS-Munich.
Ref.: Guido J. Wasser, Wasserwerk, Elektor-Plus 4, 1986, page 56 ff
Attachments
Last edited:
I do not believe for one second the claim of higher damping factor is factual. Same for improved sound quality (true only if the relay contacts are damaged).
What a Mosfet actually does is remove a service item. In my experience, something like this may get close to the performance of the original part, but it is never better. It can only be equal at best. To be honest, the only thing that performed better for sound quality was a direct connection to the speaker with a shorting element to ground for protection (called "crowbar protection"). This does normally short outputs, blow fuses and I have seen it remove large areas of copper PCB traces.
So far any mosfet type connection I have tested has added distortion to very high quality amplifiers. For an average amplifier they may be good enough these days. If I ever test one that does perform as well as a relay, I will report it, and I'll also install them in my own equipment. Haven't seen this yet.
What a Mosfet actually does is remove a service item. In my experience, something like this may get close to the performance of the original part, but it is never better. It can only be equal at best. To be honest, the only thing that performed better for sound quality was a direct connection to the speaker with a shorting element to ground for protection (called "crowbar protection"). This does normally short outputs, blow fuses and I have seen it remove large areas of copper PCB traces.
So far any mosfet type connection I have tested has added distortion to very high quality amplifiers. For an average amplifier they may be good enough these days. If I ever test one that does perform as well as a relay, I will report it, and I'll also install them in my own equipment. Haven't seen this yet.