Deep Blue Delay Buffer dull the high end. Is it norml?

[nguitar12]

So I built a Deep Blue Delay according to the following diagram:



I found that the dry signal has lose tiny bit of the high end and tiny bit of volume (Yes just a tiny bit, almost noticeable). Just wonder if the is normal and how can it be improved?

Also it this the best approach to make a input buffer or if there a better option (like fet buffer?). I want my tone as transparent as possible.

Many thanks.

[induction]

Looks like the circuit has pre-emphasis/de-emphasis. The volume of the dry signal is boosted a bit with the first op-amp (which functions like a input buffer with gain), and then attenuated at the output with the second op-amp. This is done to keep the SNR high, but you can play with the gains to get what you want. The relevant components for your issue are R3, R4, R5, R6, C16 and C20. The resistors set the gains, and the caps reduce gain for higher frequencies.

For more dry volume, try increasing R3 or R5, or decreasing R2 or R4. For more treble try reducing C16 or C20. There are other ways to do it, as well, but this will probably do the trick for you.

[nguitar12]

Looks like the circuit has pre-emphasis/de-emphasis. The volume of the dry signal is boosted a bit with the first op-amp (which functions like a input buffer with gain), and then attenuated at the output with the second op-amp. This is done to keep the SNR high, but you can play with the gains to get what you want. The relevant components for your issue are R3, R4, R5, R6, C16 and C20. The resistors set the gains, and the caps reduce gain for higher frequencies.

For more dry volume, try increasing R3 or R5, or decreasing R2 or R4. For more treble try reducing C16 or C20. There are other ways to do it, as well, but this will probably do the trick for you.

Thanks for you information. I will try this out tomorrow.

[amz-fx]

Remove C16. You don't need it and it is limiting hi freqs.

regards, Jack

[anotherjim]

Humbucking pickups? Really long guitar cable?
C16 should give a noticably cut beyond 9Khz, so shouldn't really hurt guitar signal as not much over 5Khz leaves a guitar.  But it starts to cut around 5Khz and given the wide value tolerance of caps it could be doing harm. Personally I'd drop it to 22p, as I just like the idea of having some rf rejection early in a circuit.
       
The input impedance is low for a guitar effect. Single coil pick-ups with a fairly short cable can be quite happy down toward 100k impedance and this design is higher than that at a tad under 180k. Low input impedance will reduce the output level and disproportionately cut highs from passive guitars which is why 1M is usually the ideal input impedance.

I think I know why this is, but it couldn't really be helped without increasing part count.
Since the output amp has to be a mixer for the wet/dry blend it has to be an inverting amp. To maintain overall signal polarity, the input amp is therefore also inverting. If you raise the gain setting resistors of the input amp, but keep their ratio, then you'll also raise the input impedance without changing the gain. The harm from this is an increase in resistor noise. That said, you might get useful improvement without too much noise by making R2=470k and R3=1M *PROVIDED THAT* these are metal film resistors. C16 must then either be no more than 22p (7Khz cut) or deleted. Most guitars pick-ups are happy into around 500K impedance provided the cable isn't stupidly long.

Probably a better solution is to add a High Z buffer on the input. JFET source follower would probably give the lowest part count/footprint.