Aria AD-10 Repair : help !

[buisson]

Hello everyone,
I know, it's not a DIY... but I don't know where to find help for this one !

This is my good old Aria AD-10 Analog 2 stage Delay.

Symptoms : The LED turns on, but no effect in the signal (on both stages).
Bypass OK.





I can open it, to picture the inside, or, please tell me where I can find an Expert to talk to for this kind ok repair.

[Mark Hammer]

Given your description, three things should be considered:

1) The bias could be off, such that the BBDs are not passing any signal.

2) The switching FET is shot and needs replacing.

3) The one/two-stage switch has seen better days and is not connecting the delay path to anything.

All BBD chips rely on a bias voltage at their input or else they don't pass signal.  If the unit has any sort of locking substance painted on the bias trimmer, don't touch it.  If not, consider playing with it, but save it for last.

Most delay and modulation pedals that use electronic switching (and it is rare that a pedal with a foot treadle for bypassing doesn't) will use a single FET to enable/disable the wet path, with the dry path always "one".  The circuit for turning that FET on and off can still be functional (including the LED) even though the FET itself may be shot.  Chances are pretty good that FET is a 2SK30A (the 2S will likely be omitted on it).  It is possible to test this out by simply bridging the source and drain pins with a temporary hunk of wire.  If that restores your delay signal then the FET is the culprit.

The switch may be harder to discern.  I'm assuming the pedal has 2 BBD chips in series and the switch simply takes the output of the first or the second.  Ideally, you'd identify the path from the first one to the switch, and either scope it, or use a piece of wire to directly connect it to what would be the switch's output.

[buisson]

I'm not sure of witch one you want me to test.

Is it the first transistor that the signal should pass through ? (circled in red on my first image)





Waiting for your answer, i'll try to play with those bias...


Merci ! 

[Mark Hammer]

Don't play with the bias just yet.  At least not until you know which one IS the bias.

With 4 trimmers, mistakes can be made,  I suspect one of the trimmers sets the clock range for the MN3102, another sets the maximum feedback/repeats (up to the point of runaway oscillation), a third sets the balance between the two complementary BBD outputs to minimize noise, and the fourth sets the bias.  If you were able to figure out which was which right away, you probably would not be asking the questions you did.  I don't mean that as an insult.

It is possible that they used more than one K30A in the circuit, given how many I see in the photo.  One of the transistors will probably be a 78L05 regulator (the MN3205 likes to have 5V), and two will likely be used for the flip-flop used for switching.  Another 3 or 4 will probably be C1815 or similar, used for the lowpass/anti-aliasing filtering.  So give each  transistor-shape a gentle bend backwards, if needed, to see if there is one labelled K30A.  If none are, then at least let us know what the numbers are, in case one is an acceptable sub for the 30A.

It seems my first guess about the pedal was incorrect, and the 1-2 stage switch may simply select between ranges for a single BBD (as the write-up here gives clues to: http://effectsfreak.com/effect/aria/ad-10_analog_delay/ ).  That actually makes troubleshooting easier.

[buisson]

I did manage to find one K30A (in the center of the main board).
It seems OK : 644 on both sides.

I'm managing to test every Transistor. I found a C1815, witch is close to the "Out", scoring 1327 on one pin and only 500 on the other pin. Is it a problem ?
Do all the Transistors should have the same value on both legs ?

[thermionix]

Man that thing is chock full of original Tube Screamer parts.

644, 1327, 500...what do these numbers refer to?  DC millivolts?

[Mark Hammer]

I don't know what the 644 refers to either. The critical thing will be diagnosed if you bridge the two outer pins of the K30A, temporarily, with a piece of wire.  If you get delay signal back again, then that's your culprit.

[buisson]

I've just short-circuited the K30A as you proposed.

I still have the unmodified signal from my guitar... 

PS : the values I've mesured are taken with my multimeter in "diode" position.

[Mark Hammer]

  So we know it's not that.  The next step would seem to be verifying that the BBD is properly biased.  The bias will be provided via a trimmer connected to pin 7 of the BBD, probably through a fixed resistor around 100k or so.

The bias needs to be set right or else the chip won't pass signal.  Don't worry about hurting anything by playing with the trimmer.  Sweeping it from one end to the other won't damage anything.

If you don't have a scope, I generally recommend tuning it by ear, while listening with headphones to a steady-state tone, preferably with the clean signal cancelled or at least the delay signal maxed.  That way, you get the best picture of how clear or distorted the delay signal is.  That said, whatever you can do to provide the least irrelevant sounds is good.

If the chip and bias circuit is functioning properly, then there will be a range of adjustment, both above and below the optimal bias, where you'll hear nothing then a distorted delay signal, then gradually cleaner, and finally back to distorted then nothing.

Let us know if that does something useful, and get back to us.

[buisson]

I'm sorry to ask but what is a BDD ? And witch one is it ?

Thank you Mark for your patience !

[antonis]

I'm sorry to ask but what is a BDD ? And witch one is it ?

It's a Bucket Brigade Device..
(like MN3205 & MN3102 in your case..)

https://en.wikipedia.org/wiki/Bucket-brigade_device

[buisson]

Thanks guys, I'll look into it, and will get back to you.