My wonderful preamp, which I have been using daily for around four years, recently developed a problem with one channel dropping out. You could turn the volume up and the signal would "punch through" and it would work again. Well I was poking around and I realized that when you turn the speaker EQ off it works. There is a relay that simply bypasses the whole circuit in the "off" position. When I switched the EQ back on it worked fine.
I tried to design the circuit so the signal doesn't go through any relays but this was practically unavoidable. I also had a relay fail that shorts the output for mute. It is protected by two 23 ohm resistors per channel. It has to discharge capacitors through the resistors so the failure wasn't a surprise. I installed another sacrificial relay and it works.
I'd have to sort through my notes but the relays are miniature DPDT made by Omron. They have silver and gold contacts. They are intended for low level signals (datasheet clearly states "for hi-fi"). They are obsolete but I have a couple more. I also bought a small stash of low signal level relays (a lot smaller) but I haven't used any of them yet.
So I'm hoping to hear success stories with small signal relays. What kind did you use, any tips, etc.
I'll dig up the datasheets for the relays referenced if need be. Thanks.
I tried to design the circuit so the signal doesn't go through any relays but this was practically unavoidable. I also had a relay fail that shorts the output for mute. It is protected by two 23 ohm resistors per channel. It has to discharge capacitors through the resistors so the failure wasn't a surprise. I installed another sacrificial relay and it works.
I'd have to sort through my notes but the relays are miniature DPDT made by Omron. They have silver and gold contacts. They are intended for low level signals (datasheet clearly states "for hi-fi"). They are obsolete but I have a couple more. I also bought a small stash of low signal level relays (a lot smaller) but I haven't used any of them yet.
So I'm hoping to hear success stories with small signal relays. What kind did you use, any tips, etc.
I'll dig up the datasheets for the relays referenced if need be. Thanks.
Also, this relay cycles every time the unit is turned off or on when the switch is in the "EQ" position. I always leave the "EQ" switch on.
These relays don't like being soldered more than once, either. I soldered a new one into a prototype. When I removed it I checked it out. It was FUBAR. One leg wouldn't switch. So they're expensive (around $6 IIRC) and delicate. I avoid using them at all costs but sometimes it's unavoidable. I'm building adjustable high pass filters and an adjustable biamp crossover for a high power amp and there will be lots of relays. All adjustments will be made with small signal, on board logic driven relays. I require precision and reliability and potentiometer based adjustable filters won't do what I need. I'll eat my soldering iron if fussy relays foil me.
These relays don't like being soldered more than once, either. I soldered a new one into a prototype. When I removed it I checked it out. It was FUBAR. One leg wouldn't switch. So they're expensive (around $6 IIRC) and delicate. I avoid using them at all costs but sometimes it's unavoidable. I'm building adjustable high pass filters and an adjustable biamp crossover for a high power amp and there will be lots of relays. All adjustments will be made with small signal, on board logic driven relays. I require precision and reliability and potentiometer based adjustable filters won't do what I need. I'll eat my soldering iron if fussy relays foil me.
Last edited:
I have a Yamaha C-70 preamp with relay problems like that, and I used reed relays (small signal only) as replacements. I had to make little mounting interposers to fit them, but oxidation will never occur in the hermetically sealed reeds. Preamp works fine. I used 24V coils, which is what the originals used.
Well thank you. I looked at reed relays but will have to look again.
Based on my hypothesis, I tried a procedure in an attempt to "burn off" the oxidation. I turned off the power amp, turned the volume to full, and turned the EQ switch off and on at least 200 times over the course of a couple minutes. Then I turned the volume down, turned the power amp back on, and it seems to be fixed. It made a difference in the sound. When the sound came back on, it was muffled in the right channel. Then it would drop out again. I knew I would have to fix it. After my experimental procedure, the "soundstage" has returned. So I'm virtually certain that the relay is the culprit.
So the sweet sounds are back- for now. I'll likely have to build a new board because it's so small and populated on both sides. I'll have to see if I left room to change the relay.
Based on my hypothesis, I tried a procedure in an attempt to "burn off" the oxidation. I turned off the power amp, turned the volume to full, and turned the EQ switch off and on at least 200 times over the course of a couple minutes. Then I turned the volume down, turned the power amp back on, and it seems to be fixed. It made a difference in the sound. When the sound came back on, it was muffled in the right channel. Then it would drop out again. I knew I would have to fix it. After my experimental procedure, the "soundstage" has returned. So I'm virtually certain that the relay is the culprit.
So the sweet sounds are back- for now. I'll likely have to build a new board because it's so small and populated on both sides. I'll have to see if I left room to change the relay.
Small signal relays need to be bifurcated quad crossbar contacts with gold flashed coating over silver.
Contact material by itself is not adequate at very low levels.
But the fix is simple. Most small signal relays in a plastic case have a small plastic knob on top. This is to be clipped off after the relay is soldered in place. That allows enough air to reach the contacts to prevent many aging issues.
But the simplest fix is Deoxit contact cleaner and then the appropriate lubricant.
Contact material by itself is not adequate at very low levels.
But the fix is simple. Most small signal relays in a plastic case have a small plastic knob on top. This is to be clipped off after the relay is soldered in place. That allows enough air to reach the contacts to prevent many aging issues.
But the simplest fix is Deoxit contact cleaner and then the appropriate lubricant.
Thanks to all.
Can't picture it but that's good to know. If I can get some Deoxit in there that's great.
It makes me hesitate to build something with a lot of relays. It could turn into a nightmare down the road.
Can't picture it but that's good to know. If I can get some Deoxit in there that's great.
It makes me hesitate to build something with a lot of relays. It could turn into a nightmare down the road.
https://www.mouser.com/ProductDetail/Panasonic-Industrial-Devices/TQ2-12V-3?qs=bpFJJ1fyfoDyKfBYwXiTGw%3D%3D I'll get 10 next time I order. These look exactly like the relay in question.
It's incidents like this that make me avoid selling this preamp. People have offered to buy it on the spot. I've taken it to people's houses and hooked it up to their expen$sive hi fis and they liked it better than their minimalist "hi end" preamps. It has tone controls that actually work without mucking up the sound. Almost every commercial tone control I've heard (except for the tone controls on my vintage Luxman) sounds terrible, terrible, terrible. How easy is it to make a practical bass and treble control that doesn't distort? Real easy. 😉
It's incidents like this that make me avoid selling this preamp. People have offered to buy it on the spot. I've taken it to people's houses and hooked it up to their expen$sive hi fis and they liked it better than their minimalist "hi end" preamps. It has tone controls that actually work without mucking up the sound. Almost every commercial tone control I've heard (except for the tone controls on my vintage Luxman) sounds terrible, terrible, terrible. How easy is it to make a practical bass and treble control that doesn't distort? Real easy. 😉
Last edited:
This does not occur with quality sealed "telecom" relays. I never experienced a failure with quality relays like those made by Takamisawa/Fujitsu/NAIS/Panasonic if they are used to switch voltage (and they have bifurcated quad crossbar contacts with gold flashed coating over silver like simon7000 pointed out). It is an entirely other matter when they are used to switch current.... Desoldering should also not lead to issues. The fact that they fail in 2 different scenarios says the relay brand/type is not one to recommend 🙂
So I think a better equivalent should be searched. To make it more efficient it would be nice if you could mention the exact type number of the specific relay.
edit: I see MarcelvdG recommends Panasonic TQ2. These are fine indeed.
So I think a better equivalent should be searched. To make it more efficient it would be nice if you could mention the exact type number of the specific relay.
edit: I see MarcelvdG recommends Panasonic TQ2. These are fine indeed.
Last edited:
These are indeed sealed telecom relays. I thought they would prove reliable.
I'm going to order some of those Panasonic relays. I'll need a handful for my project.
It's interesting that it's worked without a hitch since I did my procedure.
I'm going to order some of those Panasonic relays. I'll need a handful for my project.
It's interesting that it's worked without a hitch since I did my procedure.
Is that the same rule that says I won't be able to get it to screw up when I put it on the bench?
That is probably Murphy at work. When audio is concerned I just know that I haven't had a single A branded quality telecom relay fail when used correctly. You just have the odd exception apparently as the type failed in 2 scenarios. When looking way back in time I do recall relays to fail but then I mean very old relays from the seventies etc. with so so contact materials or when switching considerable power. Then small mini files were used to clean contacts 😉 A habit that unfortunately stayed in the heads of many that today sand or file gold plated contacts...
A few years ago I found relays in a building from 1979 that were used to switch on lighting and these were now starting to fail while having switched factors higher current than they were originally rated for and this for decades .... so it can be the other way around too.
But .... I did see many relays with gold plated contacts fail in audio when used to switch currents like for speaker protection. The simplified rule is to use gold contacts for voltage and silver for current.
A few years ago I found relays in a building from 1979 that were used to switch on lighting and these were now starting to fail while having switched factors higher current than they were originally rated for and this for decades .... so it can be the other way around too.
But .... I did see many relays with gold plated contacts fail in audio when used to switch currents like for speaker protection. The simplified rule is to use gold contacts for voltage and silver for current.
Last edited:
Well that's an interesting rule of thumb.
I'm not sure that using one of these relays on the output, shorting to ground for mute, is a correct application. I'm thinking that a more general purpose relay might work better. It would be bigger, which means I'll have to make a new control board.
It is not only interesting, it is also true 🙂 If one has repaired many eighties era Japanese amplifiers one is familiar with this. Always the speaker relay with gold contacts that is defective.... to be replaced for a relay with hard silver contacts... Please look up the theory behind it yourself.
Shorting to GND is normally no issue when the standard 100 Ohm stopper resistor is used before the contacts. The standard 100 Ohm resistor to prevent the circuit from capacitive loads that is. The same 100 Ohm resistor that also stops RF from coming in. These 100 Ohm resistors perform at least 3 duties but yet they are all too often omitted.... When done right the relay contacts are the last things to the RCA connectors, providing real 0 Ohm when muted so absolutely no sound and no DC (at power up/down) at all.
These tiny gold (plated) contacts are allergic to shorting high level DC to GND without current limiting with eh... let's say a 100 Ohm resistor 🙂 The 4th purpose of that 100 Ohm resistor.
Shorting to GND is normally no issue when the standard 100 Ohm stopper resistor is used before the contacts. The standard 100 Ohm resistor to prevent the circuit from capacitive loads that is. The same 100 Ohm resistor that also stops RF from coming in. These 100 Ohm resistors perform at least 3 duties but yet they are all too often omitted.... When done right the relay contacts are the last things to the RCA connectors, providing real 0 Ohm when muted so absolutely no sound and no DC (at power up/down) at all.
These tiny gold (plated) contacts are allergic to shorting high level DC to GND without current limiting with eh... let's say a 100 Ohm resistor 🙂 The 4th purpose of that 100 Ohm resistor.
Last edited:
I use 23 ohm resistors, one on each leg. 46 ohms isolates the output from the outside world. The relay has to short the power amp's input cap through 23 ohms, and the preamp's output cap through 23 ohms, simultaneously. Perhaps some on board adjustments can be made.
Should be made obviously. You will solve one of the reasons why the relay fails. Normally the present RC input filter in a well designed so RF protected power amplifier provides at least a 1 kOhm series resistor.
It would help (as always) if you have a drawing as a drawing solves problems the fastest way.
It would help (as always) if you have a drawing as a drawing solves problems the fastest way.
Last edited:
I used NEC EA2 12 nu relays to switch sources in my preamps and they work great, no failures got them from mouser. They are gold plated and sealed in inert gas.
I put them in some ic sockets to keep heat off of the pins.
Ive got 5 preamps, out inbthe world using them with about 7 or 8 in each preamp.
Highly recommend you buy from a real distributor like mouser or digi key. Gold plate on a sealed relay seems like the place the counterfeiter might skimp, if you did indeed obtain counterfeit parts
I put them in some ic sockets to keep heat off of the pins.
Ive got 5 preamps, out inbthe world using them with about 7 or 8 in each preamp.
Highly recommend you buy from a real distributor like mouser or digi key. Gold plate on a sealed relay seems like the place the counterfeiter might skimp, if you did indeed obtain counterfeit parts
Well thank you to all.
I know the relays are genuine. I only buy from Mouser, DigKey, and Parts Express. Parts Express is just for speaker stuff.
I find it quite interesting that the relay has performed without a hiccup. But there is another minor problem. The bass control has developed a slight scratchiness. I remember that the offset for the bass control circuit varied between -4 and +12 millivolts as you turned the pot from lock to lock. The pots are Bourns pots like you might buy as a replacement part for a classic piece of equipment. I will build a better circuit with smaller pots (one third the size and seem to work just as well) and much less offset. It's a modular design and I still have the same header in stock so it will be easy to change. The bass control circuit is the output, which is why I put a 5532 in there. I will put a dedicated output board in at the same time. I have some DIP-8 BUF634s in deep storage that I will have to flush out.
I know the relays are genuine. I only buy from Mouser, DigKey, and Parts Express. Parts Express is just for speaker stuff.
I find it quite interesting that the relay has performed without a hiccup. But there is another minor problem. The bass control has developed a slight scratchiness. I remember that the offset for the bass control circuit varied between -4 and +12 millivolts as you turned the pot from lock to lock. The pots are Bourns pots like you might buy as a replacement part for a classic piece of equipment. I will build a better circuit with smaller pots (one third the size and seem to work just as well) and much less offset. It's a modular design and I still have the same header in stock so it will be easy to change. The bass control circuit is the output, which is why I put a 5532 in there. I will put a dedicated output board in at the same time. I have some DIP-8 BUF634s in deep storage that I will have to flush out.
Last edited: