My amplifier (a modified Quad 405-2) produces a slight pop on power on, and a soft thump from the woofers when powered off. This isn't much of an issue, but I'd prefer to have totally silent operation.
My solution is to add a relay between the amplifier output and the speakers, which will turn on after the amp and off before the amp. I'm fine with the control side, and now just need to choose a relay.
I would prefer to use Farnell to source the parts from.
The amplifier is 100W max, so 5A max per channel, however it may make sense to get bigger relays so they're never close to their limit.
I'm looking for a DPDT relay in order to switch both channels, and a coil voltage of 48v DC.
So far the best two I have found seem to be:
- Omron 5A DPDT 48v
- Finder 15A DPDT 48v
Out of these, I'd be tempeted to go for the Omron as I've heard slightly more about them in audio - although would the 15A one be a better choice?
Or is there another relay I haven't considered? The list of possible relays Farnell sell (with the correct coil resistance, contact voltage, current and configuration) is this list - showing nine options.
Many thanks for any opinions!
Andy
My solution is to add a relay between the amplifier output and the speakers, which will turn on after the amp and off before the amp. I'm fine with the control side, and now just need to choose a relay.
I would prefer to use Farnell to source the parts from.
The amplifier is 100W max, so 5A max per channel, however it may make sense to get bigger relays so they're never close to their limit.
I'm looking for a DPDT relay in order to switch both channels, and a coil voltage of 48v DC.
So far the best two I have found seem to be:
- Omron 5A DPDT 48v
- Finder 15A DPDT 48v
Out of these, I'd be tempeted to go for the Omron as I've heard slightly more about them in audio - although would the 15A one be a better choice?
Or is there another relay I haven't considered? The list of possible relays Farnell sell (with the correct coil resistance, contact voltage, current and configuration) is this list - showing nine options.
Many thanks for any opinions!
Andy
The 15A would be a better choice.
(Would you consider a solid state relay (to diy it ?). I'm a total convert to these now. Dead easy and with same footprint as those in your link)
If you want mechanical though I would perhaps look at more substantial relays. Big issues are degradation of contacts if switching when there is significant signal present ie its turned up load 🙂 and audible degradation at low levels which is what I have experienced.
(Would you consider a solid state relay (to diy it ?). I'm a total convert to these now. Dead easy and with same footprint as those in your link)
If you want mechanical though I would perhaps look at more substantial relays. Big issues are degradation of contacts if switching when there is significant signal present ie its turned up load 🙂 and audible degradation at low levels which is what I have experienced.
As an electronic engineer, my instinct tells me to use mechanical relays rather than solid state. With solid state, the signal has to travel through active devices, which are rarely linear devices. With a relay, you've just got a few bits of metal pressed together - it seems a much simpler and reliable solution.
The 15A relay (which is actually 15A peak, 8A continuous) has a minimum switching current Edit: switching load of 5mA. That seems pretty low - is that going to affect the audio at low volumes?
Thanks,
Andy
The 15A relay (which is actually 15A peak, 8A continuous) has a minimum switching current Edit: switching load of 5mA. That seems pretty low - is that going to affect the audio at low volumes?
Thanks,
Andy
Last edited:
Isn't the switching current what the coil needs to close the contacts? I would not expect any influence on audio, as long as the DC supplied to the coil is clean.
That is just to keep the contacts clean. Switch the amp on/off with some signal, and there won't be a problem.
As an electronics engineer 🙂 that was always my view too... that you can't beat a good mechanical contact.
But think small as well as large signal. That tape hiss on a recording ! the faintest sounds ! What amplitude might they be voltage wise... how much current do they cause to pass through the relay contact. It's a lot less than 5 ma. A typical symptom of mechanical relays (as witnessed by many threads on here) is someone complains of amplifier distortion but finds it "goes" when they turn the amp up loud. And the caue is always the same... a slightly tarnished relay contact.
(so last mention of SS relays then, and a link to this rather long thread,
http://www.diyaudio.com/forums/solid-state/191449-output-relays.html
particular interest are the posts around the 200 to 300 mark)
Pull some data sheets on the latest FETs and see how low the RDs value can be. And it's consistent and guaranteed.
guaranteed for 1000's, or 10000's, or 100000s, or 1000000s, or 10000000s of operations?
A mechanical relay won't be guaranteed to switch DC currents above it's DC rating for many operations !!!!
Now I know why we (I) prefer to use the k & M multipliers.
If this is for PA use with a lot of volume (voltage) copper contacts might be okay. Down to about 1 V signal, silver plated contacts are in order. Below 1 V, you want precious metal contacts (gold, platinum, rhodium, osmium).
paralleling a precious metal contact telephone relay with a big high rated copper contact might be okay for variable service.
I'm monkeying with 40A rated copper contact motor contactors to disconect the PS rails on my amp on detection of a speaker or output transistor (DC) fault, but since they are repairable, I'm going to take the contacts out and put silver solder on them.
You might find is cheaper to re-engineer your amp to get rid of the thumps and pops. I'd start buying CL-60 or bigger NTC resistor (GE) in series with the power transformer if it is a transformer type power supply. Final numbers depend on the rating of the amp. My PV 1.3k amps is a bear of a driver, but it doesn't pop or thump.
paralleling a precious metal contact telephone relay with a big high rated copper contact might be okay for variable service.
I'm monkeying with 40A rated copper contact motor contactors to disconect the PS rails on my amp on detection of a speaker or output transistor (DC) fault, but since they are repairable, I'm going to take the contacts out and put silver solder on them.
You might find is cheaper to re-engineer your amp to get rid of the thumps and pops. I'd start buying CL-60 or bigger NTC resistor (GE) in series with the power transformer if it is a transformer type power supply. Final numbers depend on the rating of the amp. My PV 1.3k amps is a bear of a driver, but it doesn't pop or thump.
Last edited:
Working on the idea that levels under 5mA may be distorted:
- 5mA into 8 ohms is 40mV
- 5mA at 40mV is 0.0001W
From the above back-of-an-envelope sums, it seems there will usually be more than 5mA flowing so distortion shouldn't be too much of an issue. I'll give it a go with the 15A relay and report back how it sounds.
Andy
- 5mA into 8 ohms is 40mV
- 5mA at 40mV is 0.0001W
From the above back-of-an-envelope sums, it seems there will usually be more than 5mA flowing so distortion shouldn't be too much of an issue. I'll give it a go with the 15A relay and report back how it sounds.
Andy
It will sound fine. It's at some time later (weeks/months/years) that it may play up.
You see its happened to me.
Amp in a clean environment and clean air so this is the kind of problem affects other folk not me.
Right... err... wrong 😀
I wasn't aware audibly (on music) that there was any issue but playing a sine wave of around 400hz (from memory) and increasing the volume and one channel was horribly distorted. I went back to music and coudn't fault it. Weird ! I really thought my speaker was faulty. I turned the volume up more and more and it suddenly cleared. It was the relay.
More of the same here. Just two of many that I could remember.
Posts #8,9 and 10
http://www.diyaudio.com/forums/solid-state/216999-kenwood-ka-94-power-trafo-specs.html#post3112287
http://www.diyaudio.com/forums/solid-state/215483-help-warm-up-noise.html
Andy,
It is not so much that under 5mA there will be distortion, but rather that the relay wants at least 5mA of current while switching in order to keep the contacts clean.
vac
I can only echo what Mooly has said. I never expected my relays to start acting up, but that's what they did and after only a couple of years. The main issue with standard relays though is that they aren't particularly good at switching high DC currents and it's DC that your amplifier will produce under a fault condition. This DC is what will fry your loudspeakers and if the relay fails to open, because the contacts have welded shut due to the DC arcing, then your relay didn't do it's just properly and you're down a pair of speakers AND a relay.
The solid state relay on the other hand is perfectly suited to switching large DC currently, it is after all what lots of MOSFETs are designed to do in the first place anyway. These will ensure that your loudspeakers are protected and they will also switch faster then a mechanical relay too.
I posted measurements in the thread that Mooly linked to showing the performance of the amplifier with the bad relay and then measurements after with the solid state relay in place, with the problem eliminated.
The solid state relay on the other hand is perfectly suited to switching large DC currently, it is after all what lots of MOSFETs are designed to do in the first place anyway. These will ensure that your loudspeakers are protected and they will also switch faster then a mechanical relay too.
I posted measurements in the thread that Mooly linked to showing the performance of the amplifier with the bad relay and then measurements after with the solid state relay in place, with the problem eliminated.
My 2 cents worth:
Use paralleled contacts - as many as you can. 4PST/DT can be had,
more is better. Assuming it is in series with the amp output, not shunting to ground. Although more there is not a bad idea either.
Can't speak to the SS relay shunting to ground.
Single contacts? For me, never.
Ymmv.
_-_-bear
Use paralleled contacts - as many as you can. 4PST/DT can be had,
more is better. Assuming it is in series with the amp output, not shunting to ground. Although more there is not a bad idea either.
Can't speak to the SS relay shunting to ground.
Single contacts? For me, never.
Ymmv.
_-_-bear
I suppose the point would be, that if you've got multiple contacts with dual pole relays, with the 'open' state being the output of the amplifier shorted to ground, then you only need one of the relays to successfully open, because even if the other ones weld shut, one has at least shorted the amps output and protected the loudspeakers.
I still don't see why this is superior to a solid state affair, where you can connect them so they are either floating, or ground referenced. Heck you could use a mechanical relay + solid state where the SSR, with it's faster switching time, is used solely to connect the output of the amplifier to ground during the presence of a fault.
I still don't see why this is superior to a solid state affair, where you can connect them so they are either floating, or ground referenced. Heck you could use a mechanical relay + solid state where the SSR, with it's faster switching time, is used solely to connect the output of the amplifier to ground during the presence of a fault.
My amplifier already has DC protection, so the relays are just to connect the speakers just before power on and disconnect them just after power off.
I've installed the 15A relay into the amp and it currently seems excellent, although it will be interesting to see how it changes over time. I've used crimp on bolt terminals to connect it, so swapping it out for a different relay should be very easy if I wish to try that.
Andy
..and that should have read "connect just after power on and then disconnect them just after power off".
The amps capacitors take about 10 seconds to discharge before there is the thump, so having the relay turn off 2 seconds after the power is removed works fine.
Andy
The amps capacitors take about 10 seconds to discharge before there is the thump, so having the relay turn off 2 seconds after the power is removed works fine.
Andy
I'm using a simple RC circuit wired as a low pass filter (connected to 50v ac via a single rectifying diode) to provide the delays. The maths, if anyone's interested:
- Series resistor: 270R
- Coil resistance: 3.4k
- Thevenin resistance: 250R
- Capacitance: 470uF
Which leads to:
- Power on delay = ~0.13 seconds (250*470E-6)
- Power off delay = ~1.6 seconds (3400*470E-6)
Which is ideal to avoid both power on (which happens immediately) and power off noise (which happens when the PSU caps discharge below 15v and the opamp switches off, so 8 to 10 seconds after pulling the plug).
- Series resistor: 270R
- Coil resistance: 3.4k
- Thevenin resistance: 250R
- Capacitance: 470uF
Which leads to:
- Power on delay = ~0.13 seconds (250*470E-6)
- Power off delay = ~1.6 seconds (3400*470E-6)
Which is ideal to avoid both power on (which happens immediately) and power off noise (which happens when the PSU caps discharge below 15v and the opamp switches off, so 8 to 10 seconds after pulling the plug).
Last edited:
- Status
- Not open for further replies.