Re: Re: Loudspeaker Relays
Fotios,
You may have a very good point here. Not only may the contact closing force be increased, but the excess force may cause some slight deformation in the contact arm (if it is of a springy material) that will result in some contact wiping action on closure.
Cheers,
Bob
fotios said:
Fotios,
You may have a very good point here. Not only may the contact closing force be increased, but the excess force may cause some slight deformation in the contact arm (if it is of a springy material) that will result in some contact wiping action on closure.
Cheers,
Bob
Re: Re: Loudspeaker Relays
Hello Fotios,
I think the max voltrage for a relay is more like the voltage that will guarantee that it pulls in with the specified reaction time. It isn't really a max voltage in that anything above it will damage it. It will get a bit hotter, but in my experience this isn't very critical.
jan
fotios said:
Hello Fotios,
I think the max voltrage for a relay is more like the voltage that will guarantee that it pulls in with the specified reaction time. It isn't really a max voltage in that anything above it will damage it. It will get a bit hotter, but in my experience this isn't very critical.
jan
myhrrhleine said:
Hi, thanks! I still use that THD analyzer.
I agree, my guess is that distortion created would be negligible when the Triac is open. But I should test it. I guess that even if it measures negligible there may be people who assert that it might cause some sonic degradation that is not measurable by conventional means.
Cheers,
Bob
PEAVEY CS SERIES OUTPUT PROTECTION SCH.
For the record only, i quote bellow the upgraded protection circuit that used Peavey in its latter version of CS series amplifiers in which incorporated also a relay. In the early CS series it used only the TRIAC (to short the out in gnd in the case of dc presented) driven by the bidirectional switch SBS14 (MBS4992). This is the classic crowbar named protection. The dissandvantage of this method as i remembered, was that if one output transistor shorted and the main fuse was not blown immediatelly, then one or more transistors also failed unjustly and in the most cases also the TRIAC. To resolve this problem, Peavey in its new CS series then added also the relay by keeping also the crowbar circuit. The important thing in the scheme it is the smart connection of SBS after the output contact of relay. By this way the protection of speaker it is doubled and also the output transistors and the TRIAC saved from failure caused from one only shorted transistor. In the case of DC presented in output, the SBS activated and the TRIAC triggered immediatelly to short the output in gnd saving thus the speaker from destruction. But after half a second, the relay open its contact and thus the SBS deactivated and stop to trigger the TRIAC. Thus in the same time the speaker remains disconnected and the short of output to gnd also stopped. If one wonders why Peavey perserved yet in the use of crowbar as the presense of relay it is enough by alone, the reason it is that this company it is uncommonly reserved and carefull in comparison with other companies as for the strength of its products. I remembered well in a seminar that this point was emphasized particularly from the lecturer of the company. He said that "our products first of all are builded to be durable yet under the worse conditions".
Fotios
For the record only, i quote bellow the upgraded protection circuit that used Peavey in its latter version of CS series amplifiers in which incorporated also a relay. In the early CS series it used only the TRIAC (to short the out in gnd in the case of dc presented) driven by the bidirectional switch SBS14 (MBS4992). This is the classic crowbar named protection. The dissandvantage of this method as i remembered, was that if one output transistor shorted and the main fuse was not blown immediatelly, then one or more transistors also failed unjustly and in the most cases also the TRIAC. To resolve this problem, Peavey in its new CS series then added also the relay by keeping also the crowbar circuit. The important thing in the scheme it is the smart connection of SBS after the output contact of relay. By this way the protection of speaker it is doubled and also the output transistors and the TRIAC saved from failure caused from one only shorted transistor. In the case of DC presented in output, the SBS activated and the TRIAC triggered immediatelly to short the output in gnd saving thus the speaker from destruction. But after half a second, the relay open its contact and thus the SBS deactivated and stop to trigger the TRIAC. Thus in the same time the speaker remains disconnected and the short of output to gnd also stopped. If one wonders why Peavey perserved yet in the use of crowbar as the presense of relay it is enough by alone, the reason it is that this company it is uncommonly reserved and carefull in comparison with other companies as for the strength of its products. I remembered well in a seminar that this point was emphasized particularly from the lecturer of the company. He said that "our products first of all are builded to be durable yet under the worse conditions".
An externally hosted image should be here but it was not working when we last tested it.
Fotios
Hi Jan
What is the contact resistance of the relay you described?
I've wondered about the relay too. I use the lowest resistance contacts relay I can find, (currently I use an Omron type) because it seems to me that the relay could spoil the damping effect if it raises the output impedance of the amp.
Whether a few tens of milliohms of the relay makes a difference on a normal speaker compared is a question perhaps is of little practical concern, but there again, it could be worse with a relay
...unless the feedback path includes the relay.
In which case perhaps a medium value resistor is needed across the relay terminals in case the relay is opened....
which is my current thought. Using a 10k feedback resistor and a 1k 4W resistor across the relay terminals allows the amp. to remain in control if the relay opens and provide good damping when closed...
cheers
John
What is the contact resistance of the relay you described?
I've wondered about the relay too. I use the lowest resistance contacts relay I can find, (currently I use an Omron type) because it seems to me that the relay could spoil the damping effect if it raises the output impedance of the amp.
Whether a few tens of milliohms of the relay makes a difference on a normal speaker compared is a question perhaps is of little practical concern, but there again, it could be worse with a relay
...unless the feedback path includes the relay.
In which case perhaps a medium value resistor is needed across the relay terminals in case the relay is opened....
which is my current thought. Using a 10k feedback resistor and a 1k 4W resistor across the relay terminals allows the amp. to remain in control if the relay opens and provide good damping when closed...
cheers
John
Hi Bob,
My first thoughts would be, that I would be surprised if a triac caused any audible effect, but thinking about this, has anyone ever measured the junction capacitance of a large power triac, and more importantly does it vary with applied voltage ? I don't know. It would have to fitted after the output inductor to avoid upsetting the stability of the amp.
Regards Karl
My first thoughts would be, that I would be surprised if a triac caused any audible effect, but thinking about this, has anyone ever measured the junction capacitance of a large power triac, and more importantly does it vary with applied voltage ? I don't know. It would have to fitted after the output inductor to avoid upsetting the stability of the amp.
Regards Karl
Hi Fotios,
Have been looking at your circuit of the Peavey "triac protection scheme", it's interesting. Was there an integrating network on the gate of the triac ? and if so wonder what the time constants are and also what the triggering voltage of the diac is. Once triggered the triac will remain in conduction maintaining a short on the output until the holding current is reduced to near zero, such as when an output or supply fuse blows.
Regards Karl
Have been looking at your circuit of the Peavey "triac protection scheme", it's interesting. Was there an integrating network on the gate of the triac ? and if so wonder what the time constants are and also what the triggering voltage of the diac is. Once triggered the triac will remain in conduction maintaining a short on the output until the holding current is reduced to near zero, such as when an output or supply fuse blows.
Regards Karl
john_ellis said:
Hi John,
I don't know the contact resistance, sorry.
Your idea to take the fb after the relay is basically a good one but has some issues. When the relay opens, the far end where the speaker is attached will go to ground (because of the low speaker impedance) even with a 1k across the contacts. So the fb signal goes to zero, and the amp will be heavily overdriven; not a healthy situation.
In my ec amp (see the Bob Cordell thread on error correction) I can do this trick, because the amp forward path only has an ol gain of 20...
jan
I don't see relays in the output signal path as a big issue, provided the caveats concerning contact type and construction (see Self) are adhered to. As for damping factor effects - nonsense - the cable has more resistance and what about the cross over (where in many cases there is an inductor in series with the woofer)? no point in obsessing about relay contact resistance when its only one of a number of factors (and maybe not th e biggest one).
Wrt to contact arcing or burning of one of the contacts etc, its best to follow the advice of a previous post on this thread and mute the input first, then operate the output disconnect relay. A simple pair of signal relay contacts in parallel with the input of the amp will do the trick. I normally turn my pre-amp off first, or turn it down so the speaker realy contacts always operate under ZVS mode.
I use 3 x 16A double contact relays in parallel to switch the speaker outputs (so total of 6 contacts per channel). I dig the sound when they switch in: a nice resonant 'Klunk'.
Wrt to contact arcing or burning of one of the contacts etc, its best to follow the advice of a previous post on this thread and mute the input first, then operate the output disconnect relay. A simple pair of signal relay contacts in parallel with the input of the amp will do the trick. I normally turn my pre-amp off first, or turn it down so the speaker realy contacts always operate under ZVS mode.
I use 3 x 16A double contact relays in parallel to switch the speaker outputs (so total of 6 contacts per channel). I dig the sound when they switch in: a nice resonant 'Klunk'.
john_ellis said:
Hi John,
I would worry less about the DF than about the distortion that nonlinear contact resistance could cause. A fairly large 80 m-ohm contact resistance would limit the DF to 100 even if the amplifier proper had an infinite DF. However, the voltage drop across that contact corresponds to fully 1% of the signal. If the contact reistance itself had 10% distortion, that would result in 0.1% distortion at the amplifier output.
Cheers,
Bob
G.Kleinschmidt said:Have those who worry about relay contact distortion ever considered using a relay to switch the supply rails of the power output stage instead of the speaker output?
Yes, I have, and that has been a tempting idea. I also have considered using MOSFETs to switch the rails. One just needs to choose carefully the way in which the rail voltages are controlled so that unexpected current flows and damage do not occur when the rails are opened. If this is done right, it is probably better than rail fuses because one can then make sure that both the pos and neg rails are disabled at the same time.
The only concern is that one then loses the other advantage of the relay in that it can be timed to eliminate turn-on and turn-off thumps that some amplifier arrangements can create.
Cheers,
Bob
Mooly said:
Hi Karl
Sorry for my omission. You are right about the integrating network and the time constant. To avoid the redrawing of the circuit, the omitted device it is a non polarized electrolytic cap of 2,2ìF/50V which of the one terminal it is connected between the 47K resistor and the SBS, and its other terminal it is connected in the Gnd. I hope to this give an answeer in your query.
Regards
Fotios
Bob Cordell said:
Bob,
I personally subscribe to the "dragster" protection approach!
That's nitro fuel and big tires, thousand or so horsepower, drop the hammer when the light turns green!
If the engine blows up halfway down the strip... oh well... take it into the shop and rebuild it
You folks can tell me if any of these various protection schemes have saved your speakers or not? Perhaps they do these days, but the last time I was in a position to have big amps, playing big speakers loud, no protection circuit could do anything other than prevent fires... seriously. Like closing the barn door after the critters are out...
I suppose that for consumer gear you have to provide some sort of "protection" scheme to keep various agencies and the like happy.
Quite frankly, I don't like rail fuses either. (for testing , sure...)
Silicon fuses seem to work fine for me?
Perhaps the triac works... dunno...
As far as the sonic effect of the untriggered triac hanging off the output... wonder if you can string them in series or not... is it measureable or not? In theory it should be a very high Z when off, but who knows what it looks like for sure WRT higher order harmonics or the like at <100dB down with big AC floating around it... would be good to know.
I'm not an authority by any means on this topic, so this is just my current orthodoxy.
_-_-bear
bear said:
Bob,
I personally subscribe to the "dragster" protection approach!
That's nitro fuel and big tires, thousand or so horsepower, drop the hammer when the light turns green!
If the engine blows up halfway down the strip... oh well... take it into the shop and rebuild it
You folks can tell me if any of these various protection schemes have saved your speakers or not? Perhaps they do these days, but the last time I was in a position to have big amps, playing big speakers loud, no protection circuit could do anything other than prevent fires... seriously. Like closing the barn door after the critters are out...
I suppose that for consumer gear you have to provide some sort of "protection" scheme to keep various agencies and the like happy.
Quite frankly, I don't like rail fuses either. (for testing , sure...)
Silicon fuses seem to work fine for me?
Perhaps the triac works... dunno...
As far as the sonic effect of the untriggered triac hanging off the output... wonder if you can string them in series or not... is it measureable or not? In theory it should be a very high Z when off, but who knows what it looks like for sure WRT higher order harmonics or the like at <100dB down with big AC floating around it... would be good to know.
I'm not an authority by any means on this topic, so this is just my current orthodoxy.
_-_-bear
From the other side Bear;
My Mission780 have repaired woofers - from me - because my nephew is very careless and with his Rotel 2X100W he has broke the terminal wires of woofers exactly in the point where are soldered with the slight voice coil wire. The local agency in Greece sell the spare 5" woofers for 100Euros per piece. For 200Euros total, i preffered to destroy a little the spider so i can to solder again the broken join in each speaker. I don't know if you are millionaire, but i know with certainity that i am not.
Fotios
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.