Ampeg Flanger - Stuck bypassed, any ideas?

[DaveM]

I just bought a Ampeg A-5 Flanger off of evilbay.  It doesn't work. LED lights up and turns on and off with the footpedal, but with the effect on it sounds no different than off, regardless of any of the settings on any of the knobs, or how where the single trimpot inside is set.  Schematics are not to be found.  All I can find on is this forum page here > http://www.wojnygitarowe.pl/printview.php?t=857&start=0&sid=fa83742d6ee95ade907a158480f043ea , which google claims is in polish, but can't translate it.  I think from googles crude translating that that guy has the same problem as me.  Any help would be appreciated.

EDIT: After some much-appreciated translating by ptkw, it seems that link was of little use.

General stuff about the pedal:
Has 2 NE4558 dual op amps
1 MN3006 BBD
1 MN3101 Clock Driver
6 2SC1815 NPN transistors
1 K30A GR (obsolete FET I think)
1 MA522R (which is claimed to be a wierd diode here >http://archive.ampage.org/threads/1/fxgd/043419/Ampeg_A-5_flanger_part_search-1.html)
2 1S1588 diodes

According to DiscoFreq's effect database, this pedal is the same as the Guyatone PS-004, Rozz R-5 and Westbury W-04.

What I don't understand is how the buffered bypass functions with one, maybe two FET's.  I always though it required at least three, one to switch the bypass, and two for the effect.

The seller say he'll take it back at my shipping expense, or give me a discount if I keep it. I'd like to fix it though. Any ideas?

[ptkw]

I can't give you any help beyond translating the Polish thread for you. You guessed right that the original poster had the same problem as you (the effect powers on, but the output signal isn't modulated). Unfortunately, there's no definitive explanation or solution given to his problem. Halfway the thread he says he observed that there isn't any output signal coming from MN3101 clock; to that someone answers that he should probably try replacing said IC.

[DaveM]

Wow, that was quick! Thanks, I guess I'll try and see if mine has any clock signal.

[ptkw]

I'm very glad I was able to help. Being a total beginner in electronics, I thought that I have registered on this forum only to ask nagging questions. Luckily, it seems that I have some helpful skills after all, i.e. I'm a native Polish speaker.

Please keep in mind though that the original poster never confirms whether the MN3101 was the root of the problem at all or whether he managed to repair his pedal. For all I know (and I know virtually nothing), no output voltage on the IC might be perfectly all right. You should probably wait for some verification from one of this forum's experts; perhaps there are some other possible causes of your problem.

[DaveM]

I'm not exactly familiar with BBD chips either, and from what the internet has told me about them, I might have to pull the chip to test it. Which would be a shot in the dark. If it is broken, then I have to source a new one, which seems to be easier said than done, as there are fakes galore out there.  Yep, I'm definitely going to wait to see if any of our in-house experts have an idea or two.

[oldschoolanalog]

Serious troubleshooting demands voltage readings (for starters). Do not pull anything just yet.
Let's start w/the clock & BBD and work from there.
The single K30A is most likely the switching FET. Lots of effects w/buffered bypass use only one FET for that function.
V's first. Then (if you don't already have one) make yourself an audio probe. You will need to check the audio signal at various points of the circuit at some point. Also, hunt down a schematic if you can.
What did you pay for this if you don't mind me asking?

[Mark Hammer]

A person would need a lot more hands and feet to count the number of commercial flangers that use a single FET to "bypass" switch.  With many kinds of modulated effects (phaser, chorus, flanger), all that is needed to make the effect go away is mere cancellation of the wet signal, which only requires a single JFET.  The dry signal still passes through all the usual stages on the way to the output, and is fed to the BBD.  The only thing that is different during bypass is that the wet/delay signal never gets fed to the mixing stage that combines dry and wet.

The simple thing to do is to jumper the JFET by soldering on a piece of wire on the copper side where the K30A is.  The datasheet for the transistor says that this will be pins 1 and 3.  That will make the unit behave as if there was no switching JFET, and in theory you should hear the effect again.  If that does not do the trick, then the problem lies elsewhere.

[DaveM]

Mark, jumpering the outside pins of the fet did not work.

oldschool, I paid $20 US + $15 shipping.  Haven't made an audio probe yet, but when I do, where should I begin probing? Anyways, here are your voltages.

Voltages (1-8)
MN3006
1 9.3
2 8.8
3 4.25
4 9.15
5 0
6 8.8
7 8.9
8 8.9

MN3101
1 9.15
2 8.8
3 0
4 8.8
5 8.8
6 8.7
7 8.6
8 9.15

NE4558N
1 5
2 5
3 5
4 0
5 5
6 8.2
7 8.7
8 9.4

NE4558N
1 Varies between 1-7 depending on position of width pot
2 4.8
3 4.8
4 0
5 4.8
6 4.8
7 same as 1
8 8.36

K30
1 8.62
2 0
3 8.3-effect on 2.94-effect off

All measurements were taken with all pots at max.

[Govmnt_Lacky]

Right off the bat, it looks like your LFO (The 4558 chip that is oscillating) is not working fully. I believe that it SHOULD be osscillating on Pins 1, 2, 3, 5, 6, and 7. You only show it oscillating on pins 1 and 7.

Also, something does not look right with the Op Amp (The other 4558) on pins 6 and 7. I think those two pins SHOULD be around 5V as well.

[DaveM]

Here's some audio probe results:
Oscillator 4558
1 Loud Tick
2 Tick
3 Loud Tick
4 No sound, or hum
5 Tick
6 Tick
7 Tick
8 Nothing

Other 4558
1 Boosted audio
2-8 Nothing

K30a Mosfet
1 VERY quiet audio
2 nothing
3 Audio

Does this help any without a schematic? I've found nothing after repeated searches.  The only schematic I've found for these series of Ampeg pedals was the distortion.

[Mark Hammer]

Hmmm, you didn't by any chance happen to buy it from a guy named Chuck, did you, the author of this post? http://archive.ampage.org/threads/1/fxgd/043419/Ampeg_A-5_flanger_part_search-1.html

The good news, sort of, is that if the BBD is shot, you can pop in an MN3007, or MN3009, in the same spot and it should work.  My guess is that the MN3101 will likely need to have the clock cap value dropped to achieve the same delay range, but it should work.  If you happen to have anything else or can borrow something, that uses an MN3007, you can pop the MN3006 in the same socket to test it out.  Given that the 128-stage MN3006 is probably clocked a little slowish to achieve an appropriate delay range, the lowpass filtering may be a tad low to keep the noise out.  But that's just a guess not an actuality.

I can't find anything regarding the 3-pin diode.  Do we have any indication that it works or doesn't?

The usual recommendation given the age of the pedal (minimum 25 years) is that the electrolytic caps likely need some replacement.

[DaveM]

*fanfare plays*
I contacted the Guyatone Tech guys, who led me to the Guyatone Japan guys, who found me a schematic! *imitates Bubbles from Trailer Park Boys* "Decent!"

http://www.instructables.com/files/orig/FT8/RQYS/GE38T59V/FT8RQYSGE38T59V.pdf

Everything appears to match the Ampeg, save for their dual Op-amp choice (mine has 4558, they have 1458), including the wierdo three pin diode, which is used in the switching circuit and has two of it's pin directly connected anyways.  Diode testing shows it works, it reads .8 on my DMM on way and open the other.  That is voltage drop right, .8V?

I do have a Chorus Pedal with an MN3101 and MN3007, so I'm going to try swapping them out tonight, hopefully that will work.  But in the meantime, we now have a schematic.

[Mark Hammer]

Excellent!

Nothing particularly exotic about the design, though you will note the use of passive dry/wet mixing (R16/R19).  You will also note that R6 (180k) is noticeably less than R7 (120k), suggesting this is not going to be a high resonance monster when you finally get it working.  You may wish to tinker with R6 and maybe drop it down to 150 or160k.  And if you want to be able to dial back the flanging a bit, R16 might be replaced with a 10k resistor and 25k pot.

I draw your attention as well to R25/C19.  Those two perform a nice trick which was one of my favourite parts of the first issue Small Stone.  They form a single-pole lowpass filter, with a rolloff commencing around 1.6hz.  What this does is provide some "rounding" of the LFO waveform as you move from slowest to fastest speeds.  Think of it as adaptive waveform shaping.  A nice feature to have, and so simple.

Certainly the use of a 128-stage BBD and a clock cap vaue of 27pf (C20) suggests this thing sweeps darn near zero, but isn't going to sweep particularly low int he spectrum.  That will certainly make it great for bass, though perhaps less desirable for guitar.  We'll let the final outcome make the decision there.

[PRR]

> R6 (180k) is noticeably less than R7 (120k), suggesting this is not going to be a high resonance

R11R12R13 R15 give gain like 1.66. The loop gain so far is 1.1111... I see R8 R9 give gain of 0.909... It nominally gets very close to "unity" loop gain.

[DaveM]

I tried swapping out the BBD in the flanger for one in a chorus I have.  The BBD in the flanger is fine. Problem is, i bricked my chorus in the process. Thank god I only paid $20 for that thing.  Anyways, this has peeved me off too much, and I'm sending it back to the guy I bought it off of.

[g.]

Same problem here !!

changed the fet, this is not the problem

any ideas ?

thanks!