Hi !

I have read a lot of posts regarding distortion when running 100% volume out and most of the responses suggest gain staging problems.

I have really tried my best to find the cause of distortion om my XS7. It happens when volume is close or at max with certain sounds like rhodes with chorus and a lot of bottom frequencies. I have tried factory reset, switching RAM and tested the RAM using onboard diagnostics and everything seems fine but sound wise it is not.

Just to make sure this post will not get the usual “gain stage treatment” I have connected my XS Rack to the same Focusrite Pro40 device with the exact same cables and gain staging and also made sure the main output level is set to +4dB on both and also used the very same voices and played the very same chords.

The result is the XS7 making the typical distortion many others have complained about but the XS rack sounds just fine. The level of both Motif devices are similar but the distortion on XS7 is really annoying.

Is there anything one can do to fix this, changing some caps or so ? To me it sounds like the A/D converters don’t keep up when there is too much energy to deliver.

The problem is not present when using the headphone jack

I have attached a file to let you guys hear the problem. The file is recorded with 3:rd party banks and is really making the problem stand out. When using internal presets, the distortion ins not that obvious but it is still there and matches other peoples descriptions.

Any help appreciated

/Anders

MotifXS7_FullOutput.mp3  (File Size: 631KB - Downloads: 631)

Hi, what voice did you use in the recording you have submitted?

Stone

helle - 09 April 2015 08:58 PM

[...]The problem is not present when using the headphone jack[...]

Since the PHONES output signal is the same one that goes to the L/MONO and R line outputs (other than a stage of amplification), you can forget about A/D converter and similar issues.

A likely cause is leaky output muting transistors. I’ve posted about the problem several times. All you need is a DC voltmeter connected to the line outputs for verification. Do a search - if you can’t find some threads discussing it, let us know and I’ll find them.

stoneb3 - 09 April 2015 09:28 PM

Hi, what voice did you use in the recording you have submitted?

Stone

I’m not sure but it’s one of the rhodes on from the Jamal’s Flash Sounds for Yamaha Motif XS*/XF/MOXF

The site where I bought it is http://store.gospelmusicians.com/Yamaha_Motif/

Good stuff IMHO

5pinDIN - 09 April 2015 09:39 PM

helle - 09 April 2015 08:58 PM

[...]The problem is not present when using the headphone jack[...]

Since the PHONES output signal is the same one that goes to the L/MONO and R line outputs (other than a stage of amplification), you can forget about A/D converter and similar issues.

A likely cause is leaky output muting transistors. I’ve posted about the problem several times. All you need is a DC voltmeter connected to the line outputs for verification. Do a search - if you can’t find some threads discussing it, let us know and I’ll find them.

Thanks a lot

I will serch for the information and check the DC.
Think I have read about it here some time ago.

I will report back to this thread when I investigated more.

Again, thanks

/Anders

5pinDIN - 09 April 2015 09:39 PM

helle - 09 April 2015 08:58 PM

[...]The problem is not present when using the headphone jack[...]

Since the PHONES output signal is the same one that goes to the L/MONO and R line outputs (other than a stage of amplification), you can forget about A/D converter and similar issues.

A likely cause is leaky output muting transistors. I’ve posted about the problem several times. All you need is a DC voltmeter connected to the line outputs for verification. Do a search - if you can’t find some threads discussing it, let us know and I’ll find them.

From another thread: “The 2SC2878 transistors can be replaced, or the whole JK board....”

This was a discussion re: Motif Classic, Do the transistor spec apply to XS as well ?

Newer once have the surface mounted 2SC3326 according to one thread. Is “Newer” XS and XF or where is the breaking point ?

Sorry for all the questions but hopefully the thread can be helpful for others dopping by.

/Anders

helle - 10 April 2015 06:06 AM

From another thread: “The 2SC2878 transistors can be replaced, or the whole JK board....”

This was a discussion re: Motif Classic, Do the transistor spec apply to XS as well ?

Newer once have the surface mounted 2SC3326 according to one thread. Is “Newer” XS and XF or where is the breaking point ?[...]

The ES and earlier models used the 2SC2878 transistors at all audio outputs - PHONES, L/MONO, R, ASSIGNABLE L and R. Starting with the XS, only the PHONES outputs have the 2SC2278, while the line outputs use the 2SC3326, surface mounted to the underside of the board. For the XS, the board in question is the JKAN-JA.

When the muting transistors become leaky, the failure can present itself in one or more ways, depending on the degree of leakage and how the synth is being used. Possible symptoms are distortion, unwanted signal coupling between outputs, and “static” noise when operating a passive volume pedal that’s connected to the line outputs. Connecting the Motif’s output to an amplifier which has DC-coupled input could result in DC at the amp’s output. If the amp’s protection kicks in, the amp will shut off signal, but there could be damage to the amp or connected speakers if the protection doesn’t work.

This thread should provide additional insight:
http://www.motifator.com/index.php/forum/viewthread/474301/

5pinDIN - 10 April 2015 08:08 AM

helle - 10 April 2015 06:06 AM

From another thread: “The 2SC2878 transistors can be replaced, or the whole JK board....”

This was a discussion re: Motif Classic, Do the transistor spec apply to XS as well ?

Newer once have the surface mounted 2SC3326 according to one thread. Is “Newer” XS and XF or where is the breaking point ?[...]

The ES and earlier models used the 2SC2878 transistors at all audio outputs - PHONES, L/MONO, R, ASSIGNABLE L and R. Starting with the XS, only the PHONES outputs have the 2SC2278, while the line outputs use the 2SC3326, surface mounted to the underside of the board. For the XS, the board in question is the JKAN-JA.

When the muting transistors become leaky, the failure can present itself in one or more ways, depending on the degree of leakage and how the synth is being used. Possible symptoms are distortion, unwanted signal coupling between outputs, and “static” noise when operating a passive volume pedal that’s connected to the line outputs. Connecting the Motif’s output to an amplifier which has DC-coupled input could result in DC at the amp’s output. If the amp’s protection kicks in, the amp will shut off signal, but there could be damage to the amp or connected speakers if the protection doesn’t work.

This thread should provide additional insight:
http://www.motifator.com/index.php/forum/viewthread/474301/

Thanks for the clarification and the link.
I am ordering the components as we speak and will replace the leaking ones in a few weeks after checking for DC wich I most likely will find.

I really like my motif except for this issue and I got it cheap (now I know why)

/Anders

helle - 10 April 2015 08:31 AM

Thanks for the clarification and the link.

You’re welcome.

 

helle -

I am ordering the components as we speak and will replace the leaking ones in a few weeks after checking for DC wich I most likely will find.

I presume that you’ve previously worked with small surface mount devices. (For those who haven’t, I suggest that this isn’t a job on which to learn the technique - replacing the entire board or having a good technician do the work might be prudent.)

By the way, do you have the service manual?

 

helle -

I really like my motif except for this issue and I got it cheap (now I know why)

It often surprises me how a minor problem causes some people to give up on otherwise good gear…

5pinDIN - 10 April 2015 08:51 AM

helle - 10 April 2015 08:31 AM

Thanks for the clarification and the link.

You’re welcome.

 

helle -

I am ordering the components as we speak and will replace the leaking ones in a few weeks after checking for DC wich I most likely will find.

I presume that you’ve previously worked with small surface mount devices. (For those who haven’t, I suggest that this isn’t a job on which to learn the technique - replacing the entire board or having a good technician do the work might be prudent.)

By the way, do you have the service manual?

 

helle -

I really like my motif except for this issue and I got it cheap (now I know why)

It often surprises me how a minor problem causes some people to give up on otherwise good gear…

I have done some soldering and I’m prepared to give it a try, the problem I imagine when it comes to Surface mounted components is all the pins that needs to be heated at the same time both when removing and mounting them.

This chip has only 3 pins so I guess it is not that hard.

I have colleauges that have done a lot of SMT soldering so that’s my plan B

I do have the Service Manual and have changed some components in this keyboard before.

i too have little understanding for people who just give up.

To hijack my own thread, I also have a feeling I should replace the key release damper strip since I think it makes a little too much noice when playing. Is this a known issue and a likely cause ? The noice is less noticable at the first an last octave so it seems to be a bit worn out.

Best regards, Anders

helle - 10 April 2015 09:14 AM

I have done some soldering and I’m prepared to give it a try, the problem I imagine when it comes to Surface mounted components is all the pins that needs to be heated at the same time both when removing and mounting them.

This chip has only 3 pins so I guess it is not that hard.

I have colleauges that have done a lot of SMT soldering so that’s my plan B

The transistors are quite small, so that’s part of the challenge. Heating all the leads at once will help with removing the part without damaging the foils. Once it’s off, if excess solder is removed, then the replacement can be soldered without concern of heating all the leads at once. I’ve found that a bit of paste flux applied to the board before soldering helps hold the part in place and minimizes the possibility of solder bridging. Be sure to de-flux afterward with an appropriate solvent, especially if you use a water soluble flux.

 

helle -

To hijack my own thread, I also have a feeling I should replace the key release damper strip since I think it makes a little too much noice when playing. Is this a known issue and a likely cause ? The noice is less noticable at the first an last octave so it seems to be a bit worn out.

If it’s your own thread, are you really hijacking it?  ;-)

Are you referring to what Yamaha calls the “Stopper U” in the service manual? Of course, it does get extra compression every time a key is released, so if the typically more used portion of the keybed is noisier, it might help to replace it. Sighting down the length of the keyboard, and seeing if the top front edge of the white keys seem to be noticeably misaligned, could help with deciding what to do.

5pinDIN - 10 April 2015 10:30 AM

helle - 10 April 2015 09:14 AM

I have done some soldering and I’m prepared to give it a try, the problem I imagine when it comes to Surface mounted components is all the pins that needs to be heated at the same time both when removing and mounting them.

This chip has only 3 pins so I guess it is not that hard.

I have colleauges that have done a lot of SMT soldering so that’s my plan B

The transistors are quite small, so that’s part of the challenge. Heating all the leads at once will help with removing the part without damaging the foils. Once its off, if excess solder is removed, then the replacement can be soldered without concern of heating all the leads at once. I’ve found that a bit of paste flux applied to the board before soldering helps hold the part in place and minimizes the possibility of solder bridging. Be sure to de-flux afterward with an appropriate solvent, especially if you use a water soluble flux.

 

helle -

To hijack my own thread, I also have a feeling I should replace the key release damper strip since I think it makes a little too much noice when playing. Is this a known issue and a likely cause ? The noice is less noticable at the first an last octave so it seems to be a bit worn out.

If it’s your own thread, are you really hijacking it?  ;-)

Are you referring to what Yamaha calls the “Stopper U” in the service manual? Of course, it does get extra compression every time a key is released, so if the typically more used portion of the keybed is noisier, it might help to replace it. Sighting down the length of the keyboard, and seeing if the top front edge of the white keys seem to be noticeably misaligned, could help with deciding what to do.

Thanks for the soldering tips, really good advice

I will check the Stopper U to see if I can see anything indicating wear but I am quite sure there are no misaligned keys. Maybe there is just some part within that needs fastening. I could be something resonating, i think.

Needs more investigating

I talked to the authorized service center today and they said the DC leakage on fresh muting transistors should be as low as 10 mV and they said no more than 50mV was okish

/Anders

helle - 10 April 2015 10:11 PM

[...]I talked to the authorized service center today and they said the DC leakage on fresh muting transistors should be as low as 10 mV and they said no more than 50mV was okish

From one of my posts in the thread I previously linked to:
”There should be very little, if any, measurable DC voltage - if all’s OK, under 0.1 volts, and likely significantly less. While it’s possible for leaky output coupling capacitors to cause DC to appear, in my experience it has always been the muting transistors causing the problem.”

So it seems we basically agree - I was saying no more than 100mV, and typically much less.

When you check the L/MONO output, plug a cable into the R output. Otherwise, any leakage from the right channel will show up at the L/MONO jack, even if the left channel is OK.

Often the assignable outputs are OK, apparently because they don’t typically get used as much as the main line outs. It’s still worth the time to check them, since once the board is out the additional effort to replace another part or two is minimal.

Best of luck in getting this resolved.

helle - 10 April 2015 09:14 AM

...To hijack my own thread, I also have a feeling I should replace the key release damper strip since I think it makes a little too much noice when playing. Is this a known issue and a likely cause ? The noice is less noticable at the first an last octave so it seems to be a bit worn out....

I believe you have diagnosed this correctly.

See my post in This Thread.  The discussion here concerns weighted actions (BHA and GHA), but I truly believe you will find the dynamic very similar for the synth weighted actions...over (perhaps more) time.

This hammer “rebound damper strip” is not difficult to replace. 

Once done, you will probably feel like you just got a new keybed!

I know I did with my P200 piano.

cmayhle - 11 April 2015 12:13 AM

helle - 10 April 2015 09:14 AM

...To hijack my own thread, I also have a feeling I should replace the key release damper strip since I think it makes a little too much noice when playing. Is this a known issue and a likely cause ? The noice is less noticable at the first an last octave so it seems to be a bit worn out....

I believe you have diagnosed this correctly.

See my post in This Thread.  The discussion here concerns weighted actions (BHA and GHA), but I truly believe you will find the dynamic very similar for the synth weighted actions...over (perhaps more) time.

This hammer “rebound damper strip” is not difficult to replace. 

Once done, you will probably feel like you just got a new keybed!

I know I did with my P200 piano.

Thanks, I will replace it since it was not that expensive.
Hopefully it will be like new as you suggest.

5pinDIN - 11 April 2015 12:09 AM

helle - 10 April 2015 10:11 PM

[...]I talked to the authorized service center today and they said the DC leakage on fresh muting transistors should be as low as 10 mV and they said no more than 50mV was okish

From one of my posts in the thread I previously linked to:
”There should be very little, if any, measurable DC voltage - if all’s OK, under 0.1 volts, and likely significantly less. While it’s possible for leaky output coupling capacitors to cause DC to appear, in my experience it has always been the muting transistors causing the problem.”

So it seems we basically agree - I was saying no more than 100mV, and typically much less.

When you check the L/MONO output, plug a cable into the R output. Otherwise, any leakage from the right channel will show up at the L/MONO jack, even if the left channel is OK.

Often the assignable outputs are OK, apparently because they don’t typically get used as much as the main line outs. It’s still worth the time to check them, since once the board is out the additional effort to replace another part or two is minimal.

Best of luck in getting this resolved.

Did measure main and assignable tonight but before reading about the mono/stereo suggestions that are obvious now when reading about it.

Assignable was ok and the right main too. Left main (with no jack plugged into right) did measure somewhere around 537mV

I have distortions on both left and right so I wonder, can the DC leakage problem affect both left and right even if my (wrongly done) measurement showed only left having DC ?

/Anders