I recently re-capped my old SAE MK III amplifier and thought I was done with its rehab but discovered another problem. The output-protection relay contacts had become resistive and noisy. Knowing there was little to no chance of finding an exact replacement, I looked at the relay protection circuit in the service manual, and also found an online series of youtube videos describing the repair of a MK III that included replacing the output relay. In that video the original's relay coil was indicated to measure 830 ohms, which is an in-between value for a lot of currently-available 24V relays. Most are much lower, which would be a problem in terms of exceeding the manufacturer's maximum power dissipation specification (900 mW). Why do that if you don't have to? So....
I looked at a number of 24V relays from Omron but they all would require an additional series resistor capable of dissipating several watts. But the coil resistance of their 48V relay P/N LY2-DC48 is high enough @2.6K to use a 1.2K, 1W series resistor, while keeping the PD in the relay coil below the max-limit. The available drive voltage is 75V so THAT is not an issue!
It's about $14USD from DigiKey (2025 prices).
Just putting it out there for someone else now or in the future. The lifespan of industrial electrical components tends to be much longer than consumer-style devices so (hopefully) this info will be useful for some time to come.
BTW I checked the bias voltages that set the idle current in the amp's output stages. After 30+ years they are pretty close to the value given in the service manual. I can't complain about that!
A second BTW: the service manual is incorrect w/regard to the measurement points, at least in the text description...it says they are the collectors of Q13/Q14, which are the driver transistors for the output board. The schematic is correct, indicating that the points are the BASES of those transistors (but the most-convenient measurement points are the collectors of the transistors driving Q13/14 i.e. Q9/Q10). I didn't find any comments on the internet regarding this...but it has to be the case.
I looked at a number of 24V relays from Omron but they all would require an additional series resistor capable of dissipating several watts. But the coil resistance of their 48V relay P/N LY2-DC48 is high enough @2.6K to use a 1.2K, 1W series resistor, while keeping the PD in the relay coil below the max-limit. The available drive voltage is 75V so THAT is not an issue!
It's about $14USD from DigiKey (2025 prices).
Just putting it out there for someone else now or in the future. The lifespan of industrial electrical components tends to be much longer than consumer-style devices so (hopefully) this info will be useful for some time to come.
BTW I checked the bias voltages that set the idle current in the amp's output stages. After 30+ years they are pretty close to the value given in the service manual. I can't complain about that!
A second BTW: the service manual is incorrect w/regard to the measurement points, at least in the text description...it says they are the collectors of Q13/Q14, which are the driver transistors for the output board. The schematic is correct, indicating that the points are the BASES of those transistors (but the most-convenient measurement points are the collectors of the transistors driving Q13/14 i.e. Q9/Q10). I didn't find any comments on the internet regarding this...but it has to be the case.
"Updatemydynaco" has a modern replacement relay kit for the Dynaco ST400/416, same relay as the SAE MKIIIC/M. I've used them and they are simple to install just like the original relay.
I've noted the SM mistake a couple of times to people working their amps.
Craig
I've noted the SM mistake a couple of times to people working their amps.
Craig
Their replacement doesn't look at all like the one in my 3CM. It's an open-frame style and has spade type connectors on it.
It also is about $10 more than the Omron model I found so I think I will go with that one, but I will keep the one from updatemydynaco in mind if it doesn't work out -- thanks for the info!
I get the impression you've worked on the 3C/M in the past. Yes?
-Mark
It also is about $10 more than the Omron model I found so I think I will go with that one, but I will keep the one from updatemydynaco in mind if it doesn't work out -- thanks for the info!
I get the impression you've worked on the 3C/M in the past. Yes?
-Mark
I specialize in SAE, GAS, and SUMO. I've done a few strip and start over MKIIIC/M amplifiers. The Updatemydynaco relay kit mounts just like the original relay and connects the same.
Hi Mark,
Coil resistance is approximate. If it is higher - great. That is just a more sensitive relay. a little lower shouldn't hurt anything.
Several amplifiers from earlier times used relays that were called "plug-in", they could accept terminals, solder or be installed in a socket. There are so many variations, it is difficult to rememebr which amp used what relay.
Coil resistance is approximate. If it is higher - great. That is just a more sensitive relay. a little lower shouldn't hurt anything.
Several amplifiers from earlier times used relays that were called "plug-in", they could accept terminals, solder or be installed in a socket. There are so many variations, it is difficult to rememebr which amp used what relay.
It doesn't look much like mine but I haven't pulled it out of the amp yet.:
That's a photo I took before I re-capped the output board. The previous owner added the film caps. He did some other much-funkier stuff I had to un-do before I could gain access to the backside of the driver and output boards.
Here's a better photo of that old relay:
The relay was wired using solder joints, not spade connectors. My mistake there. It attaches to the chassis with two 6-32 screws, shown above.
There's enough space in that location to accommodate the Omron relay I bought to replace the old one. While I was in the process of removing the old relay I found a cold solder joint on one of the coil connections. That could have exacerbated the contact problem.....but the relay contacts do look pretty corroded. I tried burnishing the relay contacts when I re-capped the board but clearly that wasn't too successful.
Now I need to make a mount for the new relay. I have a benchtop milling machine so it won't be too difficult to make it. I'll fabricate an aluminum plate with two holes drilled/tapped for 6-32, and glue the relay to the plate. A socket for this relay costs almost $4 and would require some mounting antics as well so I don't think a socket actually would buy me much.
The relay isn't a nicety. Turning the amp on, the VU meters momentarily bounce up to their maximum. That would be a pretty nasty "thump".
Oh, and the LED replacements for the VU meters' burned-out pilot lights look quite nice. I had the choice of several colors but stuck with warm white. Blue would be a nice look but maybe later. My wife suggested finding some that change colors but even if such a thing was available, I think that would be a distraction. I don't want to re-live my college-era color organ, been there done that.....
The relay was wired using solder joints, not spade connectors. My mistake there. It attaches to the chassis with two 6-32 screws, shown above.
There's enough space in that location to accommodate the Omron relay I bought to replace the old one. While I was in the process of removing the old relay I found a cold solder joint on one of the coil connections. That could have exacerbated the contact problem.....but the relay contacts do look pretty corroded. I tried burnishing the relay contacts when I re-capped the board but clearly that wasn't too successful.
Now I need to make a mount for the new relay. I have a benchtop milling machine so it won't be too difficult to make it. I'll fabricate an aluminum plate with two holes drilled/tapped for 6-32, and glue the relay to the plate. A socket for this relay costs almost $4 and would require some mounting antics as well so I don't think a socket actually would buy me much.
The relay isn't a nicety. Turning the amp on, the VU meters momentarily bounce up to their maximum. That would be a pretty nasty "thump".
Oh, and the LED replacements for the VU meters' burned-out pilot lights look quite nice. I had the choice of several colors but stuck with warm white. Blue would be a nice look but maybe later. My wife suggested finding some that change colors but even if such a thing was available, I think that would be a distraction. I don't want to re-live my college-era color organ, been there done that.....
The new relay is installed. I didn't do myself any favors w/regard to the placement of the relay -- the wiring around the relay pins is on the tight side, and it didn't need to be. I AM a bit concerned about the 1.2K/1W resistor I added to the coil-driver circuit. When on, it will dissipate about 1/2W but the resistor is the same size as a 1/4 watt resistor! I put heat shrink around it so it may get pretty warm....
I put everything back together and, being careful, used an external current-limited +/- 40V bench supply to power it up. That was enough to get the relay contacts to close -- which means that the output-protection circuit is happy. The output offsets are a bit too high but that's not surprising, the supply rails normally are +/- 75V.
I want to check the distortion specs before I go any further, but to do that I need to finish making my measurement setup -- an 8 ohm dummy load plus some protection resistors for my UCA202 ADC/DAC. The high-impedance side will be in a metal enclosure to keep as much 60Hz out of the signal as possible.
I put everything back together and, being careful, used an external current-limited +/- 40V bench supply to power it up. That was enough to get the relay contacts to close -- which means that the output-protection circuit is happy. The output offsets are a bit too high but that's not surprising, the supply rails normally are +/- 75V.
I want to check the distortion specs before I go any further, but to do that I need to finish making my measurement setup -- an 8 ohm dummy load plus some protection resistors for my UCA202 ADC/DAC. The high-impedance side will be in a metal enclosure to keep as much 60Hz out of the signal as possible.