DR effects loop and DD20 finally sorted after 2 years

[cheameup]

So love my 2009 DR . the only complaint I have with it is I have been struggling with the effects loop and my delay for 2 years.posted here about it many times trying to fix it .

My boss dd20 was the only thing I wanted to run in it as I like all other effects in front of the amp .I wanted to run it at max wetness on the loop and I could never figure out why it was bootsting the volume with the loop on. I tried every possible knob position on the amp and setting on the DD20 including the -20 and +4 modes . if I ran the effects loop more than 50% wet I have a mid/volume boost when I turned on the loop. I had a work around but was never 100% satisifed with the results

So I was pulling apart my board last night to sell a couple of pedals I'm not really using and I thought I'd try my ibanez fl9 flanger in the effects loop just to see how it sounds. I turn the loop on with mix at 50% and theres a volume drop when it was unity before . that's weird. i turn the loop mix knob on the amp to full and bingo my effects loop is now untiy at full mix.

I have absolutly no idea why putting the flanger before the delay has fixed it . I'm guessing it has to be an impedence thing . can bad phase cause a volume/mid spike ? I tried another 2 pedals and they didnt fix it. only the flanger . the other 2 were true bypass though ,all my other pedals are , while the FL9 is buffered so I'm guessing that is why,but the dd20 is bruffered too so I dont really get it .But anyway guess I like flanger in the loop now

 

[elvis]

Yes, it sounds like an impedance thing. Many, if not most, delays are derived from studio effects that were originally designed for line-level low-impedance systems. If your FX loop send is not buffered (less than 1k Ohm output impedance), then it needs to drive into a buffer that has at LEAST 1Meg Ohm input impedance. If not, you will get an overall level drop, plus a filter effect due to the DC blocking capacitors combined with the resistance mismatch.

The FX loop on most Mesa amps (as well as others, Mesa is not a special case) is quite high impedance, often up to 25k or more. The reason is that for convenience and cost savings, the FX send is a divided-down version of the output of the preamp stage. The divider resistors are high impedance. Newer amps have either opamp or tube buffers (cathode-followers) that have much lower impedance.

To make matters worse, even with an opamp output buffer on the FX send, Mesa adds a 5k series resitor to keep people from accidentally blowing up the opamp by plugging the FX send into the wrong thing. This is a really good thing to do, but again requires some buffering.

I recommend a good low-noise buffer like the Xotic EP boost, MXR Micro amp, or even a simple JFET buffer that you can build yourself for less than $15. I have used those, plus my favorite, the Valvulator. That's a tube buffer with great sound quality and low noise. But it's $$$.

You can usually get information about input and output impedance of pedals, but many times part or all is missing. You may have to search a bit to find out.

The other thing to watch out for is that the FX return of many amps is LOW impedance. So you need a REALLY low output impedance to drive back into the amp. Again, as a general rule, I recommend FX in the loop have at least 1 Meg input impedance and less than 1k output impedance.

For the tube-buffered loops, this is much less of an issue. The rule of thumb is that input impedance should be at least 100x bigger than output impedance. So for a non-buffered 25k FX send, you want a buffer with 2.5Meg input impedance. That will be hard to find, so you may have to live with 1 Meg. For a tube-buffered loop, the cathode-follower output impedance of a 12AX7 is 600 Ohms. That will let you use a pedal with an input impedance as low as 60k Ohms. Still, more is better. On the flip side, if the FX return has an input impedance of 40k, then you will need to drive that with a pedal that has less than 400 Ohms output impedance. Again, hard to find, so you may have to live with 1k.

For those special cases, I build my own. Those can be found here: http://www.muzique.com/lab/buffers.htm
The most useful is the 4th circuit from the top. I use 2Meg resistors for bias to get a total input impedance of 1Meg. I use Toshiba 2SK170 JFETs. The output impedance is not 3.3k (the output bias resistor), rather it is the output impedance of the JFET, which is lower (I forget how much).

 

[jpdennis]

Yes, it sounds like an impedance thing. Many, if not most, delays are derived from studio effects that were originally designed for line-level low-impedance systems. If your FX loop send is not buffered (less than 1k Ohm output impedance), then it needs to drive into a buffer that has at LEAST 1Meg Ohm input impedance. If not, you will get an overall level drop, plus a filter effect due to the DC blocking capacitors combined with the resistance mismatch.

The FX loop on most Mesa amps (as well as others, Mesa is not a special case) is quite high impedance, often up to 25k or more. The reason is that for convenience and cost savings, the FX send is a divided-down version of the output of the preamp stage. The divider resistors are high impedance. Newer amps have either opamp or tube buffers (cathode-followers) that have much lower impedance.

Elvis, is the true for my Mark III, .50 Cal, & Heartbreaker? I know, probably a dumb question, but ...

Dennis

 

[cheameup]

Thanks elvis a ton of info there

the recto loop is tube buffered yeah ?


troukle shooting this loop boost problem I still had it happen with just a short patch lead across the loop with no pedals .

but for whatever reasont he flanger has fixed it and I'm super happy

 

[elvis]

Unfortunately Mesa did some bad trickery there. The FX send and return do have tubes set up for buffering (V4 A and B), but they set up the send level potentiometer after the buffer. So the output impedance depends on the send level and can be as much as 25k if the schematic I found is correct.

AND for the FX return, there is a resistor voltage divider and mix circuit that is pretty low impedance. It's pretty much a disaster if you don't use buffered pedals in the loop.

As far as the cable bridging FX send to FX return, there is a 22k potentiometer (FX mix) that is part of the resistor divider. Depending on where the mix knob is set, the internal gain between fx send and return (the parallel path) changes. Adding the external path will add additional signal by bypassing part of the mix resistor divider.

It's really a bad circuit design. What it does well is reduce the DC voltage and protect the internal circuits and external FX (and the user) from damage in case of hot-plugging or mismatched DC levels or human error (plugging an output into an output). But from a signal point of view, you have to buffer the heck out of it.

 

[cheameup]

yeah right. Its all a bit past me but wouldnt it just be so much eaisier if the loop send just controlled the volume of the loop not the entire amp ? seems that way to me .woudl be much eaisier to balance the byapssed and loop engaged levels

 

[elvis]

Maybe, but that's not what the circuit does. It will work a lot more like you want it to if you buffer your FX loop. Or just go ahead and do the series mod.

Dennis: Probably. I would have to check the schematics, but I've seen a lot of Mesa schematics and they tend to look the same.

A good buffer is your friend. It makes a HUGE difference in the tone, and allows you to set the loop for 100% wet without tone suck. The tone is actually sucked by the amp's FX loop circuitry combined with low-impedance external FX. Amp by itself: Fine. FX by itself: Fine. Amp + FX, terrible. Amp + Buffer + FX: Fine. Though this one may need Amp + Buffer + FX + Buffer.