Eventide fl201 instant flanger feedback mod
- Thread starter Seon
- Start date
Help Support GroupDIY Audio Forum:
Feedback in delay efx is typically positive feedback, so too much positive feedback can cause runaway, not enough positive feedback can lead to thin efx.
I suspect the mod is tweaking the feedback gain, more or less I can't speculate but you will easily know if you have too much. Not enough is subjective.
JR
I suspect the mod is tweaking the feedback gain, more or less I can't speculate but you will easily know if you have too much. Not enough is subjective.
JR
I tried to draw a schematic from the pictures I had.
Unfortunately I don't have picture of the back of the pcb so I can't be 100% sure of the connections between the 20K trim pot, the .001pF (filtering purpose right ?) and the 10K resistor.
I guess both pots are linear type (?).
The signal is picked up right before the last resistor of the circuit and re-injected right after the first resistor and the first capacitor before it enter the first opamp.
Does it seem correct ?
*edited my schematic, the value of the trim pot is 20K and not 10K
Unfortunately I don't have picture of the back of the pcb so I can't be 100% sure of the connections between the 20K trim pot, the .001pF (filtering purpose right ?) and the 10K resistor.
I guess both pots are linear type (?).
The signal is picked up right before the last resistor of the circuit and re-injected right after the first resistor and the first capacitor before it enter the first opamp.
Does it seem correct ?
*edited my schematic, the value of the trim pot is 20K and not 10K
Last edited:
IDK... when trying to trace out connections under a mounted PCB you can probe with a VOM to confirm trace connections underneath.
The test is how does it behave, if it runs away its too much
JR
The test is how does it behave, if it runs away its too much
JR
Hi JR.
My problem was that I only had pictures taken from above in my possession.
Finally I found a friend who showed me how he did the mod himself (not the same person who provided me the pictures).
Seems that my schematic was right but he didn't used that extra .001pF capacitor (which puzzled me) present on the pictures, also he used a 50K pot for the trim.
I post the edited schematic here if somebody needs it.
My problem was that I only had pictures taken from above in my possession.
Finally I found a friend who showed me how he did the mod himself (not the same person who provided me the pictures).
Seems that my schematic was right but he didn't used that extra .001pF capacitor (which puzzled me) present on the pictures, also he used a 50K pot for the trim.
I post the edited schematic here if somebody needs it.
Last edited:
.001 pF is likely a mistake, that is too small of a value to be effective in that circuit.
I repeat how does it sound? If it runs away its too much recirculation.
JR
I repeat how does it sound? If it runs away its too much recirculation.
JR
zmix
Well-known member
Oscillation. Feedback gain greater than 1
L´Andratté
Well-known member
Seems it would be not much different to applying the feedback externally (maybe via aux send and a 2nd mixer channel)?
What I don't get is that the piece is a flanger, and a flanger is based on feedback, if not it's just a variable delay, so there must be a feedback path already.
Are VR1, VR2 and R2 additions?
Are VR1, VR2 and R2 additions?
L´Andratté -> It might be the same if there is no input and output trans installed, because the signal is taken before the output trans and re-injected after the input trans. It was an option. Mine doesn't have trans.
abbey road d enfer -> I don't want to say something completely wrong but from my understanding of the circuit there is no feedback involved.
The signal is processed through a first BBD chip (MN3001), from there it split in two different path, one is mixed with the original signal (which went through two all pass filters) and goes to the "auxiliary output", the other one goes through a second BBD ship for more delay and then get mixed with the "original" signal which went through four all pass filters at this point and goes to the main output.
About VR1, VR2 and R2, yes, it seems like they are. VR2 being a small trimmer installed inside the gear and VR1 the control knob on the face of the unit.
I attach the (very poor quality) schematic if you want to take a look. All the bottom part is for the generation of the clock signal sent to the BBD chips.
abbey road d enfer -> I don't want to say something completely wrong but from my understanding of the circuit there is no feedback involved.
The signal is processed through a first BBD chip (MN3001), from there it split in two different path, one is mixed with the original signal (which went through two all pass filters) and goes to the "auxiliary output", the other one goes through a second BBD ship for more delay and then get mixed with the "original" signal which went through four all pass filters at this point and goes to the main output.
About VR1, VR2 and R2, yes, it seems like they are. VR2 being a small trimmer installed inside the gear and VR1 the control knob on the face of the unit.
I attach the (very poor quality) schematic if you want to take a look. All the bottom part is for the generation of the clock signal sent to the BBD chips.
Attachments
Last edited:
OK, got it. The problem is both outputs contain a version of the input signal that goes through 2 (Aux output) or 4 (Main output) All Pass Filters.
When such signals are fedback, it results in a single frequency boost that eventually turns into oscillation.
In order to fix this, you have to make sure the signal that is fedback is only the delayed signal, so either from TP3 or TP4.
Then this signal has to be injected back and there maybe no perfect solution.
Directing it to the inverting input of IC1 (via a suitable resistor) seems to be the easiest.
When such signals are fedback, it results in a single frequency boost that eventually turns into oscillation.
In order to fix this, you have to make sure the signal that is fedback is only the delayed signal, so either from TP3 or TP4.
Then this signal has to be injected back and there maybe no perfect solution.
Directing it to the inverting input of IC1 (via a suitable resistor) seems to be the easiest.
I think the proposed solution is flawed, for the reasons I exposed before. I understand they proposed it because it is easy to implement.
StephenGiles
Well-known member
You won't get a better quality schematic than that, it's hand drawn in the service manual
Thanks, it’s good to know !
