McIntosh 4100 relay replacement question

Discussion in 'McIntosh Audio' started by eedork, Sep 7, 2017.

  1. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi everyone!

    I've come across a strange issue related to the speaker protection relay (not the auto-on relay) in a McIntosh 4100 that I'm resurrecting. I'm creating a separate thread for this issue just to increase its visibility here in the forum.

    The speaker protection relay installed in my 4100 is a Magnecraft 78RCSX-11 (24VDC). I believe this is the original relay used by McIntosh since two other AKers have reported the same exact model in their 4100s. This is a 24VDC relay which I was thinking about replacing as part of my rebuild.

    While troubleshooting a separate issue I decided to measure the voltage across the relay's coil. It measured ~48VDC, which caught my attention since the installed part was only rated for 24VDC. I also noticed that the relay had condensation inside the case (see attached picture). The relay always seems to have condensation inside it, and it appears somewhat discolored in some spots.

    This discrepancy led me to the schematic, suspecting I may have a fault somewhere on the power supply board. I contacted McIntosh and obtained the correct service manual for my serial number. This is where it gets very confusing.

    The service manual shows pins 14 and 13 on the power supply board (which power the relay coil) as being 40V and 16V, respectively. This explains why a 24V relay was chosen (40V - 16V = 24V). HOWEVER, pin 14 is also connected directly to the 48V supply rail while pin 13 is connected to ground 0. So the schematic seems to contradict itself - see the attached schematic to see what I mean.

    So the question is, why was a 24VDC relay originally used when the circuit powering it is providing 48VDC? Should I use a 24VDC relay or a 48VDC relay as a replacement? I'm also amazed that the currently relay has lasted nearly 40 years if 48VDC is what has been supplied to the relay the whole time.

    Of course it is entirely possible (and likely!) that I am missing something obvious here. I've contacted a few experts and will report back once I hear from them. If anyone else has any insight, please share! Note that this 4100 now works very well, so even though the relay is being way overdriven, it appears to work. I'm just not sure what's actually correct at this point and I'm hoping to get some clarity.

    Thanks!
    -Matt
     

    Attached Files:

    Last edited: Sep 7, 2017
  2. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    FYI - these questions are still unanswered.

    For now I've decided to leave the original relay in place since it seems to basically be working. I did have one more incident yesterday where the protection circuit kicked in, so that issue has also not yet been fully resolved in my 4100. I thought I had fixed it by reflowing a number of questionable solder joints on the right channel driver board, but that does not seem to be the case. The unit had worked flawlessly for about 10 hours until it happened again yesterday.

    I need proper test equipment at this point to continue debugging. I'm looking at the Analog Discovery 2 and may purchase one in the future to help with this and future projects.

    -Matt
     
  3. dewickt

    dewickt Will fix about Anything

    Messages:
    3,389
    Location:
    Knoxville, TN
    If protection comes on to soon replace C603 the 470 uF capacitor on the power supply board.
     
    CruzBay likes this.
  4. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi Terry,

    Thanks! I've recapped the entire power supply board. The unit plays fine at high levels for long periods of time. I've had it pushing 10 Watts on the meters for over two hours with no problems.

    The issue is that sometimes when I first turn the unit on it will go into protection mode and then back out. It repeats this a few times and eventually settles down and works fine for hours. It is only the right channel that does this. Very strange issue. I thought I had it licked and was going to chalk it up to bad solder connections, but it came back yesterday.

    I have not replaced the relay because of the questions in post 1 of this thread. I have a new 24V relay on hand (which is what was installed by McIntosh) but the relay driver circuit in my 4100 is providing 48V, not 24V. The McIntosh schematic is very unclear on this.

    -Matt
     
  5. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    If anyone else in the future comes across this thread and has their 4100 apart, I'd be very curious to know what the voltage reading is across the coil of your speaker relay. I'll keep an eye on this thread and will continue to debug once I get some better test equipment.

    -Matt
     
  6. c_dk

    c_dk Addicted Member

    Messages:
    5,498
    Location:
    West Michigan
    I believe c607 on the protection section of the board can also cause issues......

    Also it would not be the first time the "sense" transistors got funky in any receivers protection scheme.

    Study the protection schematic and you will see why a 24 volt relay was used.....40-16=

    There is a 15 volt zener to regulate the voltage drop.
     
    Last edited: Sep 14, 2017
  7. c_dk

    c_dk Addicted Member

    Messages:
    5,498
    Location:
    West Michigan
    It appears you are working off the 1st schematic for the first 2000 units......there are 3 schematics, the 2nd and 3rd cover 10,000 units made.

    The power supply board and the parts layout is different but the function and voltage are much the same.
     
  8. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi c_dk -

    Thanks!

    I think the part that is confusing me is that pin 14 on the power supply board says 40V, but if you look carefully you will see that pin 14 is also directly connected to pin 1 which is the 48V supply. How can it be both 40V and 48V at the same time? This is what has been driving me nuts. It just does not make sense.

    You are correct, my 4100 is an early unit. I emailed McIntosh parts and was sent the correct service manual for my serial number, so the schematic I am looking at is the correct one.

    -Matt
     
  9. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Looks like maybe the power supply board in my 4100 has been modified. See the attached picture.

    It looks like R601 (should be 5.6k) has been replaced by a jumper. There is then a 5.6k resistor connected and what appears to be a diode (haven't pulled it to check yet) directly connected to Q606.

    Does anyone know if this was a factory modification? While recapping I assumed that it was, but perhaps it is not.

    Thanks!
    -Matt
     

    Attached Files:

  10. c_dk

    c_dk Addicted Member

    Messages:
    5,498
    Location:
    West Michigan
    What is the serial number?

    I do believe you are chasing the wrong rabbit.....

    Since an owner could freely hookup 3 pair of 4ohm speakers Mac engineering was very concerned with the heatsink heat so in this unit and the MA6200 they made the heatsink temp sensors a part of the relay circuit......overheat the unit, it shuts off the speaker load.

    I would replace the DC sense transistors to remove them from the equation and then carefully monitor the heatsink temp. It could be a bad temp sensor also but they have seemed to just fail open the very few times I have seen one fail in almost 40 years.
     
    Last edited: Sep 14, 2017
  11. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi c_dk,

    Thanks for your help!

    My serial number is BY2155.

    This 4100 plays fine except for an occasional protection glitch when first powered on, and the heatsinks are cool after an hour of playing at moderate levels. I would like to replace the speaker relay (installed part is a Magnecraft 78RCSX-11 24VDC), but my 4100 is providing 48V to the relay coil, so I don't want to replace it until I understand what is going on. Why is my 4100 providing 48VDC to the coil of a relay (the original relay installed by McIntosh) that is only rated for 24VDC?

    From the screenshot of the service manual I attached in post #1, can you tell me why pin #14 is marked as 40V but also connected directly to pin #1 which is the 48V supply?

    Thanks!
    -Matt
     
    Last edited: Sep 14, 2017
  12. c_dk

    c_dk Addicted Member

    Messages:
    5,498
    Location:
    West Michigan
    First off, I do not remember if I actually have a service bulletin somewhere or this was inside information that DOB shared at a amp clinic.

    There was a problem in the late 70s with brownouts in the Northeast parts of the US. After the blackout of '68 it was not uncommon, I understand for the northeast utilities to drop their line voltage. This drop in voltage caused the relays of the MC2205 and MC2200 to not latch correctly, owners complained to their dealers, dealers complained to Mac and so engineering come up with a solution. Which I can not put my hands on right now.....

    I think your MAC4100 has had a field modification to raise the voltage to the relay.....and the schematic shows both the original design, parts layout and field mod, circuit diagram.

    The circuit diagram seems to match your modded unit but the picture of the circuit board, both views doesn't.

    The next gen starting with BY3003 uses the board as pictured but with a circuit drawing different from yours.

    If your schematic does not say the offset detection transistors are MPS-D05......MPS-A05 today.
     
    Last edited: Sep 15, 2017
  13. clinic-audio

    clinic-audio all on YAMAHA untill 1990

    Messages:
    1,754
    Location:
    France
    Hello Matt

    Simple question : what is the DC voltage indicated on your original "blue" coil relay ?
     
  14. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi c_dk -

    Thank you for this information. If my unit does have a McIntosh mod to raise the relay voltage, would it be prudent to install a different relay? And I'm glad I'm not crazy .. it sounds like the schematic is indeed a little confusing.

    Patrice -

    The relay that is currently installed is a Magnecraft 78RCSX-11 24VDC part. I believe this to be the original part installed by McIntosh. I just removed it from my 4100 and took a picture of it which is attached. EDIT - picture removed because I circled the wrong pins!

    Thanks guys!

    -Matt

    EDIT: Please note that I circled the wrong pins in this picture! I am actually measuring across the coil pins, not the driver board pins as shown! OOPS.
     

    Attached Files:

    Last edited: Sep 17, 2017
  15. c_dk

    c_dk Addicted Member

    Messages:
    5,498
    Location:
    West Michigan
    If it has been in the receiver for almost 40 years I would not worry about it.

    Why is it intermittently cutting out should be your concern. It is easy to change the sense transistors and remove them from the equation......not so much the heat sensors although you could jumper around them if you know for absolutely positive the heatsinks are not over heating.

    Then it is just waiting to see if it will cut out again......
     
  16. clinic-audio

    clinic-audio all on YAMAHA untill 1990

    Messages:
    1,754
    Location:
    France
    Sorry Matt but since the beginning (that's what I told you long time ago by telephone) if the relay in place (original one) id a DC24V why do you think that a new one (DC24V ) will be damaged ?

    And the most important : you do not measure the coil ! there you measure the voltage between the two channels !! non sense in my point of view !

    the coil is in the green colored pin in my pic!

    Relay-4100.JPG
     
  17. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    While I generally agree that it has been working for 40 years (so don't mess with it) I really would like to understand if my unit is actually operating properly or if there is a problem somewhere on the PS board. The relay also starts to condensate, presumably from the heat of being overdriven, after about 10 minutes of use. See the attached picture. That just does not seem right.

    I'll order some MPS-A05 transistors with my next Mouser order.

    Thanks!
    -Matt
     

    Attached Files:

  18. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi Patrice -

    I am very sorry. I circled the wrong pins in my previous picture in post #14!! I am indeed measuring across the coil (the green pins in your updated pic).

    I have attached another picture where you can hopefully see the DMM and probes attached to the relay so that you can see this with your own eyes.

    This is the core issue. It is a 24VDC relay, but it has 48VDC on the coil. Is this correct??

    Thanks!
    -Matt
     

    Attached Files:

  19. clinic-audio

    clinic-audio all on YAMAHA untill 1990

    Messages:
    1,754
    Location:
    France
    very difficult to see on your picture
    is there a diode in // there ? Did you check voltage between ground and one pin of the relay ?
     
  20. eedork

    eedork Super Member

    Messages:
    1,613
    Location:
    Southern Maine
    Hi Patrice,

    Here is a better picture with more light. I assure you I am measuring across the coil, despite my previous, stupid mistake with the picture in post #12. In this picture you can see the relay 'sweating' as well.

    From the orange wire on the relay to ground is ~48VDC (this is also pin 14 on the PS board). From the purple wire on the relay to ground is 0V (this is pin 13 on the PS board).

    There are diodes on the PS board associated with the relay driver circuit. For the measurement I just have the leads of my DMM connected directly to the relay coil. D601 and D608 are the only diodes in relay driver circuit. D608 is a 15V zener that appears to be working properly since Q604+Q605 have ~15V on the collectors, as expected. I have not pulled and measured D601.

    Thanks!
    -Matt

    EDIT: I've uploaded another picture that clearly shows that there are 48VDC across the relay coil. Unfortunately I've got bigger fish to fry at this point. I'm unable to bias either channel using the procedure in the service manual (and a kill-a-watt meter) or by measuring across the emitter resistors directly. Adjusting the bias pots does not change the current reading on the kill-a-watt or change the voltage reading across the emitter resistors (0V). The emitter resistors have been tested are good. Interesting that it seems to work well with this condition - I haven't wrapped my head it yet.
     

    Attached Files:

    Last edited: Sep 21, 2017

Share This Page