Roland TR-606 hum when LEDs are on ...

[Wonks]

If it works OK with batteries, then it can't be a real problem with the TR606 and I'd imagine is down to not getting enough voltage due to the difference in PSUs.

The internal circuitry dealing with the PSU input is in the middle at the top of the circuitry of page 5 of the service manual (input shown at grid position B17).

https://www.synthxl.com/wp-content/uplo ... -notes.pdf

I don't know enough to interpret it, but someone might be able to say whether you're likely to get the full +6v,+15v and +5v with just a 9v DC input. Comments elsewhere on the ACA adaptor indicate that you're likely to get nearer 6v internally than 9v on a lot of Boss pedals.

The current draw of the TR 606 is between 70mA and 150mA depending on how many instruments are outputting at once and the volume setting.

[nathanscribe]

OK so just to clarify on the PSU requirements for the TR-606: the original Boss/Roland ACA supply was indeed an unregulated model which is mentioned in the 606 Owner's Manual. There was a later wall wart, the PSA240. This is a regulated 9V 200ma supply.

Having had my 606 for nearly 30 years, I can confirm that the LED noise you hear is to some extent one of the 'quirks' of the 606, and indeed the 303, as I had a very noisy one of those back in the day as well.

However, it is very dependent on a good supply. My 606 sounds terrible and noisy with one of those cheap generic supplies, but is perfectly fine with the official Boss PSA240. This btw is the older linear model, not the more recent switching supply version.

It also works well from a Strymon Ojai. It's very noisy from my old Maplin no-name multi-out supply.

Before scattergunning any more parts without knowing what's wrong, try a good reputable power supply.

I would hazard a guess that if its performance with batteries is clean but with a PSU it is not, there may be a grounding issue caused or exacerbated by that other PSU while your 606 is connected to other bits of gear via audio/trigs/sync.

[James Perrett]

I don't know enough to interpret it, but someone might be able to say whether you're likely to get the full +6v,+15v and +5v with just a 9v DC input. Comments elsewhere on the ACA adaptor indicate that you're likely to get nearer 6v internally than 9v on a lot of Boss pedals.

As far as I can see the power supply is regulated down to 6V which also comes direct from the battery supply. This supply is fed through a couple of diodes to give 4.8V (what Roland call 5V) and also through a switch mode circuit to give 15V. Since batteries can go below 1V per cell before they are totally exhausted, the 6V supply could easily go below 4V which would take the 5V supply down to 3.8V. I've no idea whether the TR606 will work down to those voltages but it will almost certainly work below the nominal voltages.

[tenga]

Check that there is 6V from Q42 emitter to the +6V point. i.e. across C105.
Also, the external PSU must NOT carry an earth through as that would either short out the voltage regulator, or destroy Q41, and could cause all sorts of problems as it could possibly overvolt everything!

hi Foderol and other readers.
this is what i measured.

voltage across c105 : volmeter dc 5.81v
c100 :dc 15.16v
c9 : 5.47v

i retested on batteries : the noise is also present but volume is quite smaller than with power supply.

the frequency is betwen 130 khz & 360 khz : its not a sinwave..
headphone out width vol at min


line out vol at min


this is a scope at d38 anode at the power supply.
this is the signal i hear when the led blinks....


this a redrawn version of the power supply schematic for reference


the 15v and 6v has a quite a bit of ripple : 160 mv and 80 mv respectivelly.

any hints are welcome!!

[Folderol]

Yes. That's switching spikes, so is almost certainly a voltage issue. Higher voltage input is doing something odd. Anything getting hot round there?
That voltage ratio of the spikes between them is reasonable.
C105 is a definite suspect. If it's drying out things could get pretty funky. Ideally it wants to be a high temperature low ESR type.

[tenga]

i did some testing yesterday and changed C105 and C104: no improvement.
i tried a lab power supply at 9V & 12V no improvement.
i removed D38 & D39: to check if the signal was sneaking this way: NO

probing:
actually i can see big pulses all over the place of about 100mv everyere before C105.... this means my +6V is noisy from the start.
The pulses are a bit attenuated on the +15v after the transformer.

since my +6v is noisy it sneaks into the headphone out , this is why i hear it louder on this part and a little less on the lin out wich is fed from +15v

i dont know if there is a faulty part on the power supply or if its an issue somewhere on the 6V part of the chip.

i also know the pulses are related to the Led scanning.

i'am currently looking for a thermal camera or freze spray to locate a faulty component.

any hints/suggestions are very much welcome !!
best regards
Chriss

[tenga]

Hi guys !!

any ideas based on last observation notes anyone?
i tried cold spray without any luck.
im hope i can try a heat cam soon too.

i also thought it can be one of the pushbutton that is damaged.
what do you think ?

best regards
Chriss

[Folderol]

I'm wondering if it may be something daft like a hairline fracture in a board track. Time to get the magnifying glass out. Also check for bad solder joints - especially on the bigger components.

[tenga]

I'm wondering if it may be something daft like a hairline fracture in a board track. Time to get the magnifying glass out. Also check for bad solder joints - especially on the bigger components.

this is what Daniel Araya suggested: look for hairline cracks in the pcb.
i tried with a usb microscope device but it was quite clunky...

in the meantime i tried a frezze spray without much luck: didnt see any particular hotpoint ,but i didnt had much time to look , the spray was empty in jus 1 or 2 minutes..... strange...

i also tried isopropilic alcohol to pin point any hotspot , but it melted the vernish.. so i had to stop doing it...

im pretty desperate... i will try a magnifing glass.... its quite likely to be a crack on the pcb... if you have any hints on this topic i would apreciate it

best regards
chriss

[Hugh Robjohns]

Going back to basics...

In post 1 you implied it worked perfectly when powered by batteries.

Is that really the case?

It is highly significant if so.

[jjlonbass]

voltage across c105 : volmeter dc 5.81v

This looks a bit lower than I'd expect. What DC voltages do you measure across

1. D40 (or C103, they should be the same thing)
2. R235
3. R236 (R2 in your diagram)

You say that you replaced Q42, what type of transistor did you replace it with?
Did you replace Q41? If so, with which type?

John

[tenga]

hi john!!

thanks for your reply!!!

this is what i got.

D40 (or C103, they should be the same thing): 6.507v
R235:1.980v
R236 (R2 in your diagram):3.23v

the replacements i used for q41 &q42
q42 BD242B
q41 KSA916YTA

today i looked the board with a thermal camera.
the area near q414 is a little hot.
c12 was hot also , but i removed and noise was persistent.
q42 is hot , i suppose is normal.
the are below the tempo pot is hot also.

thanks for the help once again.
chriss