Creek Destiny: Right channel thuds on switching inputs and Right channel DC Servo offset increases with volume

Hi Folks,

Any help is appreciated. I'm seeing two issues in the Right channel that seem to be related. However, without finding the root cause I cannot be certain and have listed them separately.

1. There's a slight thudding sound in the right channel upon switching inputs. The thud gets louder when switching inputs at higher volume levels. The thud is barely audible at my typical listening volumes. I can hear the thud through the headphones when the speakers are off, and through the speakers when they're turned on. I can corroborate the thuds with spikes in DC offset measurements across the speaker terminals.
  • For the left channel, I'm detecting a much much lower DC Offset spikes, and the thudding is mostly inaudible.

2. The DC Servo output (pins 8 and 9) increases when I increase the volume. For instance, the voltage from U101 (pins 8 and 9) WRT Ground = 100 mV at minimum volume (7 o'clock), and 130mV when volume is increased to 11 o'clock. DC Offset measured at the speaker terminals stays below 5mV, this tells me the DC Servo is still doing its job in spite of the voltage increase in the DC Servo.
  • The left channel DC Servo U1 (pins 8 and 9) WRT Ground = 1mV and does not change when increasing the volume.

Recent troubleshooting and work I've done on the amp
  1. Replaced the mute relay - No change.
  2. Right channel: Replaced all diodes and the IC (TL074CD) in the right channel DC Servo.
  3. Right channel: Replaced input transistors, diodes and zener diodes.
  4. Issue 1 happened before changes 1-3. I just recently stumbled on the issue 2 after changes 1-3.

Do issues 1 and 2 fall under what is considered normal for this amp? Both don't really impact my listening experience. But the concern is something is awry somewhere in the right channel from the input to the output.

Schematics attached.
 

Attachments

  • Destiny Amp R01 P4B Input Board.pdf
    280.2 KB · Views: 40
  • Destiny Amp R01 P4B Main Board.pdf
    415.5 KB · Views: 33
Last edited:

Ian Finch

Member
Paid Member
2010-04-11 4:22 am
Coffs Harbour, NSW
If the left channel still behaves normally or as you recall the original product did, then something has changed recently - probably still related to that extensive damage to the PCB assembly you showed earlier. It may not be a welcome suggestion, but perhaps your best lead would be to systematically compare the DC voltages of the input stages of both channels. As the amplifiers are DC coupled, every voltage matters but you could start with comparing the corresponding +,- and output pins of the same servo ICs in the other channel and noting the measurements with respect to the relevant power supply return or ground, on a copy of the schematic. If there is no difference, then continue with comparing EBC pins of each transistor and its opposite channel equivalent, as they occur in the input stages. You'll need to use hard point probes for your meter to avoid slips and more damage, bearing in mind that the odds get worse as you continue.

Before starting, select one pair of inputs and short their input sockets or bridge with a 1k resistor to make sure that whatever you measure down the line isn't the result of a floating potential at the input. I'd just use the 'scope to monitor the output of the preamp but you could use whatever you like to monitor the audio output.
 
Thanks for the response and method recommendation. Are issues 1 and 2 the same? The dc servo values are quite different in the left vs right channel. However, the right dc servo is apparently working fine because the dc offset at the speaker terminals is well below 10mV. Does the right dc servo output (pins 8 and 9) rise WRT volume indicate the issue is before the volume control in the Input board?
Do the voltages across R105 and R5 indicate where the issue is?

R105 = 7mV
R5= 1mV

I will send dc servo measurements later today with shorted inputs.
 
Last edited:

BSST

Member
2018-07-07 2:21 am
U1A and U101A opamp outputs apply corrections to the power stages to drive them to 0V at output. I surmise pins 8 an9 do built-i-test monitoring of system health. Right?

This is a very interesting amp. I'm still trying to find my through the innards. Are the volume controls motor-driven devices? Sheet 2 of the Input Board are the volume controls? Can you provide any details re SW501A? I think I need to understand the function entire Sheet 2 of the Input board before I can ask meaningful questions.
Your supposition re problem before volume may be correct. I'm still trying to find my way around. :confused:
 
Last edited:

Ian Finch

Member
Paid Member
2010-04-11 4:22 am
Coffs Harbour, NSW
To be honest, I can't say whether the small difference beween the resistor currents makes any odds here. The resistors load the global feedback node and the resulting voltage there represents how much total (DC+AC) feedback is being applied to each LTP. It's a global correction so you can't say whether it is correct or normal without more evidence of something, if anything in the control loop, being wrong.

Regarding the relay, I wouldn't expect a modern relay to fail yet so its some concession to have that confirmed. Issue 2 says that the signal isn't otherwise muted when the speaker relay is. That input muting will be located at the preamp output, where it controls all sources. The preamp isn't shown in the schematics you posted but it will normally mute whatever you shouldn't be hearing from any un-selected source.

I'd be surprised if the fault wasn't restricted to the right channel but you might also like to check if there is a DC problem before the power amp. by swapping pre-out and main-in channel connections at connectors JP1/101 or perhaps JP2/102 (associated with the volume control). Theyre probably keyed to prevent confusing L/R channels but there could be a simple workaround.
 
U1A and U101A opamp outputs apply corrections to the power stages to drive them to 0V at output. I surmise pins 8 an9 do built-i-test monitoring of system health. Right?

This is a very interesting amp. I'm still trying to find my through the innards. Are the volume controls motor-driven devices? Sheet 2 of the Input Board are the volume controls? Can you provide any details re SW501A? I think I need to understand the function entire Sheet 2 of the Input board before I can ask meaningful questions.
Your supposition re problem before volume may be correct. I'm still trying to find my way around. :confused:
Yeah, it took me some time before I could piece together what’s what. Thanks for the response and input. SW501A and SW501B are 3db,6db, or 9db switches for the active preamp section. I run in passive preamp mode so don’t think those are an issue.

yes volume controls are motor driven. The issue still happens when I remove the motor power.
 
To be honest, I can't say whether the small difference beween the resistor currents makes any odds here. The resistors load the global feedback node and the resulting voltage there represents how much total (DC+AC) feedback is being applied to each LTP. It's a global correction so you can't say whether it is correct or normal without more evidence of something, if anything in the control loop, being wrong.

Regarding the relay, I wouldn't expect a modern relay to fail yet so its some concession to have that confirmed. Issue 2 says that the signal isn't otherwise muted when the speaker relay is. That input muting will be located at the preamp output, where it controls all sources. The preamp isn't shown in the schematics you posted but it will normally mute whatever you shouldn't be hearing from any un-selected source.

I'd be surprised if the fault wasn't restricted to the right channel but you might also like to check if there is a DC problem before the power amp. by swapping pre-out and main-in channel connections at connectors JP1/101 or perhaps JP2/102 (associated with the volume control). Theyre probably keyed to prevent confusing L/R channels but there could be a simple workaround.
Okay, that’s good to know more data is needed in addition to those resistor values.

The preamp section is in the Input Board schematic I uploaded. I replaced the mute relay thinking that was the culprit. I verified the analogue power supplies to pins 4 and 8 in those three opamps.
I should add the issues 1 and 2 occur in active and passive mode.

The dc servo output increases with volume in the right channel. I did not notice a dc servo output increase in the left channel on the same volume adjustment. There’s also a dc offset spike when I quickly adjust the volume in the right channel.

I just remembered the right channel has more 60Hz noise! In the FFT spectrum of harmonics I sent in the other post. There’s a 6dB difference between R and L channels. That’s a 2:1 voltage ratio. Could that be the cause of issues 1 and 2? The issue is more pronounced in the right channel? And 60hz can be attributed as system noise? It seems the question is what type of issues are acceptable per design in this amp?
CDA3C7FF-A578-4DA6-ABEC-EB59D4093CBA.png
355E9E2C-A296-4A7F-8A7C-D5ABDE12682E.jpeg
 
Last edited:
Here are the dc servo voltages in the right and left channels:

All measurements are WRT ground. I verified +/- 15 volts at pins 4 and 11, and ground to pins 3,5 and 12.


U101:1 = -200mV
U1:1 = -23mV

U101:2 = -0.6mV
U1:2 = -0.5mV


U101:6 = 110mV
U1:6 = -0.5mV
U101:7 = -14.3V
U1:7 = 18mV


U101:8&9 = 120mV
U1:8&9 = -0.8mV

U101:10 = 120mV
U1:10 = -0.25mV


U101:13 = -.05mV
U1:13 = -3mV
U101:14 = 200mV
U1:14 = 20mV
 
Last edited:

Ian Finch

Member
Paid Member
2010-04-11 4:22 am
Coffs Harbour, NSW
re#8
That doesn't bode well. Connectors should still be 100% good unless they've been in a hostile environment for much of the time. If you have access to a comprehensive parts suppliers catalog like Digikey or Mouser's, you might do a search for what you need to replace at least some of them. I know, its a pain trying to replace multiway PCB connectors or fit up new cables but not impossible.
 

BSST

Member
2018-07-07 2:21 am
I do believe that the changing U101 servo voltage with increasing volume suggests that there's DC present at the PA input, i.e. JP101, pin 1. Trying comparing with same point in left channel for confirmation.

Assuming DC at JP101 is anomalous, work from there toward signal source, comparing against left channel for nominal behavior. If signal path includes U502B, pin 1, this voltage should be near 0V by virtue of servo stage U502C.
 
Thanks-I may have found something :)

I shorted the inputs to ground for these voltage readings.

At volume set to 0.
JP101 voltage = .02mV
JP1 voltage = .01mV
Both virtually 0 volts

At volume set to 3
J101 voltage = 1mV

JP1 voltage = 0.05mV

U502B and U503B pins 1 are the same around 0.02mV

So it sounds like DC voltage is being introduced into the right channel by the volume control?
 
re#8
That doesn't bode well. Connectors should still be 100% good unless they've been in a hostile environment for much of the time. If you have access to a comprehensive parts suppliers catalog like Digikey or Mouser's, you might do a search for what you need to replace at least some of them. I know, its a pain trying to replace multiway PCB connectors or fit up new cables but not impossible.
I ran across the same PCB connector (molex connectors?) issue on another Destiny that I repaired, mostly in the right channel. I think those molex plugs are bad design because mold was added during manufacturing apparently to keep a firm connection. There is a fair amount of pull from the input board on the connectors If you don’t heat up the wires ( I used a hair dryer) to relax them and not pull on the connectors. I figure since this is my main amp I should solder those connections for longevity. I also soldered the speaker connectors afte discovering a finnicky connection was the cause of noticeable channel imbalance.
 
Last edited:
the volume control checks out. I reflowed the solder, checked for any damage, and verified the left and right channel resistance matches on volume adjustment. I think the volume control is working fine and it amplifying a fault that’s from the input board. I’m focusing on troubleshooting issue #1: the thumping / voltage spikes when the mute relay is turned on and off.
It turns out: voltage spike across the speaker terminals is positive when turning the mute relay on, and negative when turning the relay off!
 
Correct, RL508 is the mute relay. Yes L andR channels have voltage spikes when muting on/off. To clarify, the right channel voltage spikes are roughly 10-20 times more than the left. At 30% volume I’m seeing 10-20mV voltage spikes across the right channel speaker terminals. I’m trying to find a way to measure the voltage at the relay wipers when the amp is powered. In the meantime, I’ve decided to hand wire all the dc servo u101 connections (there was damage to the pcb) to rule that out as a cause.
 
Last edited:
At minimal volume
The voltage across R1 and R101 stay at virtually 0 volts

When muted
The voltage across R1 and R101 stay at virtually 0 volts

At~30% volume
R101= 0.5mV
R1=.02mV (0 volts)

This happens when the inputs are shorted to ground.

Im also seeing the voltage spikes across JP101 when turning mute on and off.

R101 and R1 are surface mount resistors. I was able to measure the voltages across JP101 and JP1 while turning mute on/off when volume is set at ~30% and measured a similar voltage difference.

and the same at R520 and R519 on the input board.
 
Last edited:

BSST

Member
2018-07-07 2:21 am
I'm becoming suspicious of the right channel power amplifier: much higher bias voltage at R101 (~20 x), exercising the mute relay would give rise to clicks/thumps as the output servo drives speaker output back to null, and you mentioned having replaced some parts in this area.

I'm suspecting one of the complementary input pairs. Taking some rough estimates, I would expect there to be about 75mV across each of the four emitter resistors at the differential input transistors, R103, R104, R146, R147. You should probably check the corresponding sites in the left channel to check my sanity before probing on the suspected right channel.