Boss DD-2 Repair

[g2s]

I am attempting to repair a Boss DD-2. I work on a lot of amplifiers but this is a bit out of my wheelhouse.

I received the pedal missing D6 and the 78L05 with burn damage on the PCB. Suspect an incorrect power adapter was used.  Replaced both and now all voltages are correct.

I have dry signal but no delay. Light is not turning on (except in HOLD position, but still no effect).

Originally, I only had signal, up to IC4(Pin8). While troubleshooting I replaced IC4(4066), IC2 (NE570) and IC8(74HC00).

Putting a 100mV p-p square wave on the input, everything looks OK up to the input of IC5 (A/D converter). Coming out of the main controller, I have the attached image/signal on R73(Pin14) (1v/div). This signal does not change when I vary the input signal. My understanding is that this should be the delayed signal converted back to analog by the resistor network R73. Since this does not look right, and does not vary with input signal, I am suspecting either a faulty main controller (IC7) or a faulty a/d converter (IC5).

Any advice/pointers from those more experienced with these pedals? What is the replacement for the 271(311) a/d converter? Having issues finding data on that.

Thanks in advance for any help.

EDIT: Link to schematic: https://www.synthxl.com/wp-content/uploads/2018/04/Boss-DD2-DD3-Service-Notes.pdf

[Slowpoke101]

A 100mV square wave is not a great test signal to use. It can be a bit misleading. However it does show that the process or is working as there is a staircase pattern on the rising edge of the waveform and that also means that the R2R ladder and D to A converter is working. Try using a 20mV 400Hz sine wave signal instead of the square wave signal.
Do you still have trouble with the LED not working when the effect is engaged?

Also: Welcome to the Forum. Have fun.

[g2s]

Thanks for the response Slowpoke.

I switched input to 400hZ 20mV p-p Sine and have the exact same signal at R73(Pin14).

Now, the signal goes through IC6 and I have signal at IC4(Pin8) but nothing after that.

I still have not been able to 'engage' the effect, that is get a delayed signal on output. I also still can't get the LED to turn on.

[Slowpoke101]

Interesting result so far. For the moment it would be best to tract down the "engage" problem. Locate IC9 (BA634) and Q11. Test Q11. It is just a standard PNP transistor with no real surprises other that its pin-out. With IC9 it is easiest to test it in circuit and powered. Pin1 should show a voltage change when the foot-switch is pressed. Pin2 should show a voltage change each time the switch is pressed. IC9 is a latching flip-flop. You do not need a test signal connected whilst doing these checks.

It is good that you do get a dry signal through the unit when the effect is in "bypass" mode. That means that a lot of switching signal path is intact.

[g2s]

IC9 (BA634)
Pin 1: 9V, drops to 0 when switch is depressed, back up to 9 when let go (ok)
Pin2: 0.311VDC, not flip flopping.
Pin 3: 0V
Pin 4: 9VDC
Pin 5: 0.176VDC

Q11 passes simple diode check in circuit. Looks ok.

I'm not sure if that means bad BA634 or not?

I'm continuing to research BA634 to better understand it and see if I can find an english datasheet. Thanks so much for all the guidance thus far.

EDIT: Putting 9V signal on Pin 2 turns LED on.

[Slowpoke101]

The BA634 is not well. Pin5 is low so it's not in reset mode so it should latch without a problem. Finding a genuine one could be real fun. For the moment I suggest that you remove it and manually control the effect by applying +9V via a 1K resistor to the Pin2 connection. Then see if you get any delay effect.

[g2s]

BA634 has been removed and 9V patched into Pin 2 PCB hole via 1k resistor. LED light is consistently on.

No other change, the rest is the same: No delay signal present at output. On scope the same signal present up to IC4(Pin8), etc.

[Slowpoke101]

OK. Check what signal you have on pin7 on the 271. Also see what pin6 of IC4 is doing. Some screenshots of your oscilloscope would be helpful too.. Thanks.

[g2s]

Scope screenshots here.

These are with 20mV p-p 400hz sine on input.

I will note that nothing here (nothing after the main controller output) varies with input signal. I may be misinterpreting how the circuit works but I thought the output should change with input.

[Slowpoke101]

It has been quite some time since I've had to work on any DD-2s myself and memory does fade a bit. Unless it's clipping badly the first waveform really still should not be like that. Check what it looks like on pins pin5 and pin6 of IC2 (571). Then check pin7 of IC3. All these pin should show a reasonably nice sine wave. If it distorting, reduce the input level until it reduces somewhat. But the pedal should handle 20mV without problems.

[g2s]

Pin 6 of IC2 (570) is not connected to anything.

I think you may have meant Pin 7 - the output feeding IC3, and there is a nice sine present. No distortion.

It is still present at IC3 Pin 7 and then IC5(3). So everything going into IC5 Pin 7 looks good.

Rest is still the same.

[Slowpoke101]

Yes, silly me meant pin 7. Now, unfortunately whilst it is good that you a good signal all that way up to IC5 pin 7, thing now become very grim. You most likely have an an unhappy processor or a memory problem. Most likely a memory problem. These things use 4164 dynamic RAMs which were notorious for being prone voltage spike damage. Which this DD-2 most likely suffered when the 78L05 went bang. The 3 DRAMs are in connected parallel so it can be test which one is toast. Temperature can sometimes give it away.

This where you talk to the owner and ask how much are they willing to spend on the repair.

The processor is most likely an Hitachi 63XX series micro and they really were quite fine with voltage problems so it should be fine.

[Slowpoke101]

One more question. The VCO IC8 can be a bit difficult at times when using modern ICs. What frequencies are you getting out of this thing? Also, it may be worth try the original 74HC00 again if you still have it.

[g2s]

Looks like NTE is still making a compatible DRAM, besides that there's just various eBay sellers to roll the dice on.

re: VOC IC8, around 1.25MHz, pretty close to the original spec. I do still have the original 74HC00

[Slowpoke101]

Check out what sort of duty cycle you are getting into pin 63 of the processor. It may be suitable. If its too fast the processor won't clock correctly.
But really the problem is most likely one or more of the DRAMs. If you have the circuit diagram, check out the activity on the address lines and see if any are stuck. The same for CAS, RAS, W and D. Remember most of these are in parallel. The D line is bi-direction and depends on the state of the W line.

[g2s]

Here is the scope IC7(63). I'm not sure how to interpret it.

I took a look at address lines on DRAMS but again not sure how to interpret. Some with clean square wave, others with just a ton of noise.

I see DRAM testers are a thing and am considering purchasing one.

[Slowpoke101]

I think that the clocking signal is OK. Trying to work out if an address line is faulty just by looking at it on a scope can be difficult. Normally if a chip is causing problems it pulls the line hard low or high thereby making the problem quite easy to see. But here you may may a chip that is ignoring its incoming lines and not outputting anything. The output of the R2R ladder seems to show that the upper half of the data word is missing and the processor has just read that section as all ones hence the flat top of the waveform. Odd.
A DRAM tester, if it's cheap enough and suitable for an old style 4164 DRAM may be a good idea.

[g2s]

This DD-2 has been sitting on the shelf for the past 7 months, but I am back at it again. Hope you are still with me?

I now have a DRAM tester. I pulled all 3 4164's and tested them. 1 was bad. Replaced it with a NOS 4164 that tested good.

Still no delay audio when effect is engaged. (Currently BA634 is pulled with 9V connected to Pin 2 to keep the effect ON).

Now I have a clean sine going into IC 3 Pin (5). After leaving the first stage of the 5218, the wave appears sort of dotted (photo attached).

I have attached photos of the sine:

- Coming out of the 271, going into the Main Controller (IC 5 pin 7)
- Coming out of the D/A resistor network after the main controller and into IC 6 Pin 3 (looks very different after replacing the 4164)
- 4066 IC 4 Pin 11 (not sure if this looks right...)
- 4066 IC 4 Pin 10
- IC3 Pin 1

There is the same signal on the left side of R48 but nothing at the junction of R48 and R49. This signal varies with Delay Time knob but not with input.

Separately, I've also noticed there is no signal whatsoever at the Effect Level pot.

Any help interpreting what's happening here? Thanks in advance for any suggestions.

[PRR]

> signal on the left side of R48 but nothing at the junction of R48 and R49. ... there is no signal whatsoever at the Effect Level pot.

This is at Q10, center-right.

So what else is at junction of R48 R49? Yank C38, does it get better? (It won't be right without the filtering, but if signal passes buy a new C38.) Solder blob, stray hair-wire, bad joints?

Ultimately you MUST get signal at the Effect Level pot.

[g2s]

Ok it looks like I do have signal but it is smaller than anticipated.

Remove C38 tests fine in an LCR meter. No change.
Remove C37, tests ok. No change.

Is it an issue that the signal at Q10 base does not vary at all with input signal? It only varies with delay knob (changing the frequency).

I looked at the check points in the service manual (with 100mv 1kHZ square input as per manual) and all looks good up until checkpoint 4 which is the emitter of Q9. There should be a 2.4Vp-p signal there and there is almost nothing.

IC6(3) has 2.3Vp-p.
IC6(6) has 2.3Vp-p.
IC4(11) is 100mVp-p.
IC3(3) is 100mVp-p.
IC3(1) is 180mVp-p.
Q10(B) is 20mV p-p (approx)
Q10(C) is 10mVp-p (approx)