SA-9900 - Preamp low volume hum

Jerry's TV

Active Member
Hi all - this is a repost to start a new thread on the SA9900 ... "sa-9900-2sc1451-ksc3503-causes-dc-on-volume-potentiometer".

http://audiokarma.org/forums/index....on-volume-potentiometer.805352/#post-11288903

I used LesE's component list (thanks a bunch!!) and have worked through the power supply cards AWR-070B; AWR-077, the input equalizer amp AWK-038, switch card AWS-080, control amp AWG-031 and protection circuit AWM-0701. No changes to Tone Switch AWR-079. Rebuilding the Control Amp eliminated a “pop” with the S9 Tone on/off and clicking through the volume control pot. AWR-070 power rails are +-58V; +- 49.1; Pin 15 -12.85; Pin 17 + 12.45 (a little high).

I have not changed the main power caps or worked on the power amp yet. It seems to be performing well. I’ll do this next once the pre-amp is fully sorted.

Audio sounds good and things are working as expected… except for one annoyance. I have a subtle 60 hz background hum on both channels. It is not effected by the volume control, input selection or any other front panel controls. If I set preamp bypass switch to “bypass” for power amp only, the hum goes away. (so it is coming from the pre-amp). When I connect a multi-meter to measure the + 58V or -58 supply on power supply AWR-070B the hum will stop. Coupling issues?

Control Amp AWG-031 - If I ground the audio pins 4 and 19 input to S9 Tone Switch, I can turn the hum on/ off with tone switch; Tone Off – no hum; Tone On = hum. If I ground pins 3 and 20 after S9 the hum stops and is not switched by S9. Grounding pins 2 and 21 output of Control Amp stops the hum as well.

If the problem was Tone Switch Board AWS-079, which S9 activates, I should be able to turn the hum on/off with S9 at any time. Re-grounding pin 12 of AWS-079 does not help.
So, this leads me to think S9 Tone on/off switch is the issue? It is mounted on Switch Circuit Assembly AWS-080, schematic shows no other circuity for S9, just the switch.

Thoughts and suggestions? I am away for the week so thought I’d put this out to the smart minds on AK for discussion, then come back to it next weekend.
Brian T
 
Mark - the unit is assembled with side panels on, no bottom or top covers or cage on the power supply. I am missing the shield plate p/n ANH-233A between the control amp and the power transformer. I fashioned a piece of metal to insert in the gap as a temporary shield and gounded the metal, no change in the characteristic of the sound. I'll try putting all the covers on and see if that changes the hum.
 
Mark makes a very good point regarding the covers. I seem to recall getting some hum with the covers off so it is important to have the amp completely buttoned up before evaluating it's background noise.

Regarding AWR-070 pin 17, I belive that the 10V indicated in the SM may be incorrect. The voltage on pin 17 is used to mute the audio output of the preamp during power up. Upon turn on, Q7 is on and pin 17 is around -48V. After a few seconds, C20 charges sufficiently to turn off Q7 allowing pin 17 to swing positive and turn on JFETs Q7 & Q8 on the Switch Assembly which allows audio to pass through to the power amp. With Q7 off, the voltage on pin 17 will be set by the voltage divider formed by R20 & R21 which should be around 18V.

Next time in, I am planning on revisiting Q11 & Q12 on the control board. I haven't yet decided if I will replace R41 & R42 or go with KSC1845s with TO-92 heatsinks for insurance. The gain of the C1845 will result in a somewhat higher operating point than the 5V that you get with the C1451 but I don't imagine that it would be an issue.

Les
 
When a whole interim post I was working on for this thread recently disappeared
(did you rename the thread or delete it and start over?)
I felt like quitting. All that typing LOST. But NOT the research...

Ground the volume pot's wipers - that is the input to the circuit that you are muting the output of for testing.
By ground I mean earth, to a section of the chassis that does not have ground currents flowing through it.
Perhaps pin 10 to pin 11 which is supposed to be earthed, and pin 13 to pin 12.
OF COURSE the volume control should be set at minimum volume.
Try first with pins 11 and 12, THEN if hum, try a separate wire to a ripple free part of the MAIN chassis that is part of the star ground, if you cannot visually trace the solid metal of the sheet to the ground point, don't trust it..
Grounding is a science that too many treat as a black art. Or magic....

I will be very interested to see what that does. I am casting a gimlet eye to the biasing setup used for the current sources Q7 and Q8. They are part of the voltage amplification stage in this feedback amplifier configuration (Q1,Q3,Q5,Q7,Q9) for the
They both share a base connection for bias - a voltage divider (R25 1K to -48v reg through a series diode and R26 100K to ground.)
that is grounded, unlike EVERYTHING else in that circuit running on +48v dc reg and -48v dc reg.
 
Last edited:
Hi, this weekend i will try inserting a new shield cover between the power transformer/ control amp and replacing all covers. If not successful i'll ground the volume wipers.

Mark -- is grounding pins 4 and 19 of the Control Amp and being able to switch the hum on/off with S9 achieving the same as grounding the volume control wipers that are located further back in the circuit? (and yes i did shuffle the original post, sorry for the confusion)

thanks for the ideas.
 
Hi -- I installed the power supply cage cover, top and bottoms and sides. No change to the hum. I still need to fab a metal cover ANH-233 between the power transformer and control amp, but I have tested the control amp out, laying horizontal, as well as installed in place. I have installed a temporary shield to no effect. I'll get that shields cleaned up, but I think I still need to find the leak.

I grounded the volume control wiper at Control Amp AWG-031 pins #10, #13 - no effect on the hum.
Here is a summary of chasing the hum through the control amp...
Ground pins #4,#19 - hum will be switched on/off with S9 Tone Switch
Ground pins #3,#20 - hum is off always. output of tone amp and input of buffer amp.
Ground #10, #13 - no effect - volume control wiper input.

I also grounded pins #7& #12; #1&#2 and #11&8 of tone switch card AWS-079 - no effect to hum on any of these; with or without S9 Tone Switch engaged.
When I operate the amp on a 45 watt DBT the hum is still there, but at an acceptable level. I think regulated 58v drops to ~ 48v on this DBT.

Next step is to craft and install the missing shield, as it’s needed anyway.
Any comments welcome.
 
HI - I am a little bit closer. I found chassis grounding pin #21 of the Switch Circuit AWS-080 made a significant difference. Currently, pin #21 ground wire routes to the pre/main amp breakout connector, which is routed to the control amp grounds on pins 11 and 13. I added a new ground wire from AWS-080 #21 to chassis ground. I'm guessing this is not intended to be chassis grounded... I'm OK?

One note - chassis grounding the other ground pins on AWS-080 #14, #17, #28 did not make any difference to the hum, only Pin #21.
The ground wire from pin #21 is unsheilded and routes near the 110VAC power switch so is suspect that is the culprit.

Now when i slide a grounded shield (i use the amp's aluminum face plate) between the power transformer and the control amp for the missing shield ANH-233 - it gets better again.
Now when i ground Volume wiper pins #10 to #11 and #13 to #12 as suggested above, the hum goes away completely.

I will build my permanent shield for ANH-233 and install.
Then I'll button it back up with all covers and see where I am at.
 
Brian ... According to the schematic, AWS-080 Pin 21 should be connected to chassis ground. If you don't already have continuity between Pin 21 and the chassis, then it would be appropriate to provide it as you have done. You will also want to make sure that there is continuity between the chassis and Pin 28. Pins 21 & 28 should also be connected together through the AWS-080 circuit board traces.
 
Hi Les, I made a new ANH-233 shield and buttoned it back up. It is still there, but much better and only noticeable if you're listening for it. Low level music covers it fine.
I am going to move on to rebuilding the power amp. I'll let you know how that goes.

If anything else comes to mind to try, let me know.
 
Last edited:
Just an idea.Try a good pair of rca cables between pre-out and amp in and see if it helps. I am chasing some strange noise/ground issues on an SA-9100 and believe something is not well with all the RCA jacks.
 
Hi Les, I made a new ANH-233 shield and buttoned it back up. It is still there, but much better and only noticeable if you're listening for it. Low level music covers it fine.
I am going to move on to rebuilding the power amp. I'll let you know how that goes.

If anything else comes to mind to try, let me know.
Brian did you manage to fix this issue. I am struggling to fix exactly the same issue on SA9900 you and LesE very kindly helped me to resurrect.
I would appreciate if you could share the solution to this problem.
Thanks
Sajjad
 
Sajjad - good to hear from you. I have been meaning to get back to this thread to close it out so good you asked.

The primary issue for me was a bit of surprise, Q1/Q2 in the regulated power supply AWR-070. I replaced the new MJE15032 and MJE15033 with original devices I had removed 2SD525 and 2SB507. Humm was gone.

The way you might confirm this could be your issue too is that I could remove the humm by running the amp on a dim bulb tester (DBT) or variac, effectively lowering the main input voltage and therefore voltage to the regulators. The higher the wattage lamp i would install in the DBT, the higher the voltage, the more humm. Keep in mind i am on 115V mains... the humm was very subtle at ~ 95V, noticeable at 110V and clearly present at 115VAC. I discussed with LesE at the time we didn't arrive at any good reasons why the new parts created the hum.

Prior to finding this solution I went through all the ground connections and made sure they were in good shape. I think I outline those above. I went through each card schematic and clipped a wire from the board ground terminals to the nearest spider ground clips and if i noticed any difference i either replaced the wire or resoldered the connection. There are ground and low level signal wires for the Input card and AWR070 which run very close to the incoming power switch wiring that i found moving helped. They are on the bottom side of the amp separate the 110VAC cable routing from the other signal and DC power cables as best you can.

Another tip, I used headphones to track the humm as i was diagnosing the problem.


Let us know how you make out...
 
Last edited:
The primary issue for me was a bit of surprise, Q1/Q2 in the regulated power supply AWR-070. I replaced the new MJE15032 and MJE15033 with original devices I had removed 2SD525 and 2SB507. Humm was gone.
Interesting, thanks for posting this information.
All I can think of is that there may have been some HF oscillations going on with the higher ft bjts. It is a good idea to scope the supplies after sub'ing parts.
2sd525 is a low frequency devices with an ft of 12MHz, which is still pretty fast. imo, MJE1503x are 30MHz devices, so probably need more compensation.
This maybe a better sub list
https://www.mouser.ca/Semiconductor...0w6hvZ1z0w6bvZ1z0w6y6Z1z0w6yqZ1yyzhtoZ1yyzhp9
I guess they say one size does not fit all.
 
I have photos of the hum by using a iphone app called Spectrum and cabling the preamp output into the microphone input of my iphone.
first image is plugged into the variac but not-powered on - background level
Interesting you can see 60Hz bump just by plugging it in.
index.php


Voltage of the variac set to 95V - very low audible on headphones
index.php

Variac at 100v and humm is audible and visable
index.php

Variac set to 115vAC and this looks similar plugged into main power 115AC
index.php
 
Last edited:
I'm also getting some buzzing although to a lesser extent than Brian was experiencing. I replaced the MJE15032/MJE15033 pair with KSC2073/KSA940 which are 4 MHz devices but in my case, the change had no effect. However, I was also able to eliminate the noise when on the DBT. Using a 180W DBT, the line voltage was around 112VAC and the noise is completely eliminated. I also noticed that the power transformer was producing the same buzz when on full power but was silent on the DBT.

I'm not sure about Brian or Sajjad but I implemented the triac modification to preserve the under spec'd power switch. The next time I have the amp open, I'm planning on disabling the triac to see if it's generating hash.
 
Triacs generate hash when they are not zero voltage switched. You can see the switching on a scope.
Possible to sub a relay (120VAC coil) for the triac.
653-LY1F-AC1
I think it's unlikely that the snubberless triac (Alternistor) used in the mod is the source of the issue in my case but disabling it would be the easiest way to find out.

A relay is the best solution but space inside of an SA-9900 is very limited and mounting one would be a challenge.
 
  • I also had to replace the MJE15032/MJE15033 pair with KSC2073/KSA940 and the hum has indeed disappeared. Bundle of thanks @Jerry's TV
  • I did not try the triac mod in SA-9900 but I did it in Marantz 2270 which had started to hum. But thanks to comments above, after removing the triac from 2270, hum has disappeared.
once again I would like to thank all of you for guidance, I managed to kill 2 birds with one stone :banana:
I appreciate it
Kind Regards
Sajjad
 
I decided to open my amp today to disable the triac. Doing this seemed to eliminate the low level transformer buzz and most of the background noise.

I also checked AWS-080 pin 21 and found that it measured around 0.5 ohms to ground. Providing a solid ground on pin 21 got rid of much of the remaining hum. Great find Brian!

I also decided to shorten the timing of the preamp unmute which was occuring well after the closing of the protection relay causing a minor thump. I initially changed AWR-070 C20 from 10uf/100V to 4.7uf which was very close but still too large. Going down to 2.2uf was sufficient to ensure that the preamp is umuting before the protection relay closes. It's a minor point but I think this results in a better startup sequence.
 
Back
Top Bottom