DIYstompboxes.com

if there arent a lot of them it would probably be better and easier just to go ahead and swap them out with new ones, if you have some.

Twenty year-old caps should not be failing because of age at this point.  Many pedals that are 40 years old continue to work perfectly with caps that are twice as old.  After 20 years of being on the floor and getting kicked, there is a bigger probability that the bias voltage (set by a trimpot on the board) is incorrect.  Like any BBD chip, the MN3011 needs to have the audio signal sitting on top of a DC voltage.  If the DC bias violtage is incorrect, the  BBD will produce distorted sound.  If the bias is VERY incorrect, the BBD will produce no sound.

You can adjust this voltage by ear without needing an oscilloscope.  It is best to do this with a very clean signal and listening through a very clean amplifier/speaker with good frequency response.  You can use your guitar but it would be best to have a steady signal, like a sustaining chord from a keyboard.  Moving the trimpot around until you find the perfect setting will not damage the chip or the effect.

Two MN3011 chips on the same pedal?  I think I am going to keep my eyes open for these pedals!

Very thanks. I will take pictures to put here. I think this is a very rare pedal. Is normal the bias voltage changed? I buy this pedal this week. maybe in the future i try to make the schematic. Someone writed that its sound is like a short delay. There are 3 controls "room size" "reverb time" and "reverb level" i think.

Could someone please provide me the DOD FX45 circuit diagram/schematic
I have two MN3011(BBD) and MN3101 (clock driver) and want to make the pedal for my own.

Your suuggestion will highly be appreciated.

Thanks in advance.

Cheers,
Anshuman

Thanks Mark for your suggestion. Could you advise any specific chorus unit using MN3011, actually I managed to get two 3011 s in very inexpensive rate $3 each  and I checked  those are good. I already designed the PCB (only some confusions) after getting some good reviews for DOD FX45, though I know that it can not replace the spring reverb with maximum delay time and real effects.
If you could advise me making chorus units with these 3011 that would be really helpful for me or you have the schematic of DOD FX45.
Next i have a plan to make the digital one for my electric guitar.

Thanks once again for your reply.

Best Regards,
Anshuman

Mark... I think it's about time you got on with your stripped down A/DA! I'm still holding on to mine till a repair comes along but it could be used untill then...

Regards the FX45, why not just do a Mono Version and use one of the chips (i'm assuming it uses each per channel rather than summed and split elsewhere) the MN3011 datasheet has an example circuit and PCB layout if I remember.

If I had a MN3005 to make a Delay I would!

I heard most reverbs using this chip don't sound great... I think Mark is right about using it for Chorus rather than Reverb... It could make some kind of decent trailing or stereo delay type thing too maybe.

I have just completed etching the board, hopefully can start populating the components from tomorrow. Will update you how it is going on and of course will try to improve the performance.
Regards,
Anshuman

Thanks for the link to the schematic, Mr. Crisis.  Much obliged.

Boy, no wonder this thing didn't sell!

You will note that all the taps are mixed in equal proportion, using identical-value mixing resistors, and that all taps pass through the same filtering.  When was the last time that multiple reflection in a reverberant room all had the exact same amplitude and frequency content?

So, a noble gesture on DOD's part, but I think there was insufficient thinking put into the characteristics of actual reverb, and the circuit design reflects that.  The ETI circuit, while not perfect, is a little better in that it mixes the later "reflections" (taps) at a lower level than the early reflections (shorter taps).  Ideally, if one was going to aim for a realistic BBD-based reverb, you'd want to filter each tap, or perhaps groups of 2 to keep it simpler, separately, and then mix them together, with a little more treble taken from the later reflections/taps.  The ETI circuit mixes the taps directly, and then filters them all together.

Actually, after giving it some thought, there is probably a relatively simple way to do what I suggested.  A little bulky, but simple.

The schematic for the FX45 shows that a resistor array in a SIP provides a single 100k mixing resistor for each of the taps on the MN3011.  Lets say that the mixing resistance for the first two taps is left at 100k, but the next two taps use 1 20k mixing resistance, and the next 2 use 150k.  There ARE other resistances, and by no means are these suggested values the ideal ones.  This is just an example using standard resistor values.  Within those limitations, this would provide the suggested amplitude difference between earlier and later reflections.

But what about the differential filtering, you say?  Glad you're paying attention.   Those mixing resistances do not have to be a single resistor.  Each path could be split up into two resistors in series, with a cap to ground from their junction.  The sum of their resistances would provide the different amplitude for each tap, and the combination of the first resistor in each series pair and the cap to ground, would provide ultra-simple single-pole lowpass filtering.  So, the basic existing structure of mix-then-apply-a-single-master-lowpass-filter would remain intact; we'd just be adding a little more filtering and tweaking the mix along the way.

Let's assume that taps 1 and 2 (pins 9 and 8, at 396 and 662 stages) are left as is, with a 100k mixing resistor.  Taps 3 and 4 (pins 7 and 6 at 1194 and 1726 stages) would each run through a 68k then 51k fixed resistor, in that order, with a 1800pf cap to ground from their junction.  That's a 119k resistance in total, with a rolloff starting around 1300hz.  Taps 5 and 6 (pins 5 and 4 at 2790 and 3328 stages) pass through a 68k then 82k fixed resistor, with a 3300pf cap to ground from their junction.  That's 150k series resistance and a rolloff around 700hz.

Again, I'm not professing this would be perfect, but it is a step closer towards the ideal, providing differential mixing and progressively less treble content at longest delay times.  It should prove to be a little more realistic and provide gentler decay when recirculation is used.

Thinks I discovered. After find some cold connections, I followed the signal. There is reverbered signal (its a lot of robotic, but there is), at the input 5 of the U2, but there are no reverb signal at output of this opamp. I think this is because the non reverbered input of the opamp it's very loud, but the reverbered signal is very low. This is correct, or maybe the opamp is the problem. I think the level of the 2 inputs (clean and reverb) must be the same at the opamp input.
The transistor Q1 and Q2 are amplifying the reverb signal, but is very low than the clean signal coming from the U1 opamp.

http://www.americaspedal.net/fx45/fx45schem.jpg