Zendrive clean blend - Bass mod question

[lars-musik]

Hi fellows,

I would like to make a Zendrive more useful for a bass and figured a clean blend might just do the trick. So I searched a bit and found that nice little blend circuit:

http://kjc-guitar-betta-taro.at.webry.info/201408/img4_2.140874293520259664228.html

Now I wonder if I linked it correctly to the Zendrive schematic (from Madbeans Serendipity: http://madbeanpedals.com/projects/Serendipity/docs/Serendipity_ver.5.pdf)

I skipped the input cap from the blend circuit because the Zendrive brings its own (can i do this?) but that still makes me wonder about the other two 100n caps from the blender. Will they affect the bass frequencies?

I am thankful for all suggestions regarding zen-like bass modifications.

I underlaid the blender parts in reddish-pink.

[PRR]

I think the input buffer can also drive the Blend pot.

I think the input buffer will have higher overload and less critical JFET selection if it is biased up to VB.

The 100K Blend pot should be fed from a lower source resistance. U1.2 can easily drive 10K. We reduce R5 and increase C5 to suit.

D4 and C6 power input system will work better with some resistance to work-against. R99.

C1 is bigger than it needs to be (0.2Hz?). C101 is maybe a hair small for Bass. I just swapped them. C1 could really be 10nFd and still pass full bass.

[lars-musik]

Hi Paul,

that looks good. I'll try that. I wondered about the 470nF (C1), too. But it was in the original schematic...

Two remaining questions at the moment:

1. Do you have any advice for the JFET selection? Will any 2N5457, J201, MPF102 do?
2. Can I reduce the blend pot to 50K?

Thanks, Lars

[Groovenut]

2N5484 or the like work well for FET buffers in my experience. They seem to have pretty good drive characteristics. I would think you could use the 50k pot. The only issue being having enough isolation between signals. Should be fine though as I've seen 10k iso resistors work perfectly well is passive mixers. You'd want to modify C102 though to a 1uF to keep the same -3dB knee.

[PRR]

If you bias the JFET up to mid-supply, the specific JFET hardly matters.

The original had JFET Gate biased at ground. Source would be higher by less than Vp.The "less than" is uncertain, and Vp varies from 0.3V to over 4V. So the actual bias point, and the maximum signal level, can't be well predicted, and JFET selection may be needed.

[lars-musik]

Unfortunately this didn't work out.

I just built it and the volume is very low. All controls seems to work OK, but there is a huge volume drop and some crazy oscillations when drive and master is maxed and voice/tone controls are over 12 o' clock. Something is definitely not going as planned. I tried some different JFETS (just in case), 2N5457, 2N5458, J201.

Turning the Blend knob back, the ovedriven sound is still faintly audible - no nice and clean tones in sight,

I don't even know how where to start - so I took some voltages. Very much to my and probably your displeasure I am not a breadboard type (too many mistakes setting any circuit up). So everything that changes means cutting / bridging traces and de-soldering.
But I really want to get this going.



Ideas? Thanks in advance!
For everybody who's interested: If we can make this work, the reward will be a 1590a layout with 5 knobs and a lot of swearing during boxing it up!

[lars-musik]

There's some debugging necessary. I lifted the "out"-pin of C1, connected it directly to the "out"-pad of the (removed) C103 and grabbed the circuit output directly from the wiper of the Volume pot.

Except the changed values of C5 and R5 and the series resistor in the power supply everything should be back to stock.
Still: Massive volume drop. Are these values (C5 0.47µF to 4.7µF, R5 1k to 470R) crucial without the buffer?

More likely I made some other mistake and it is not necessarily the buffered splitter. Next time it is time to audioprobe.


By the way, is the voltage drop after the 100R series resistor in the power supply (R8/R99) correct? 0.7V seems quite a lot...

[lars-musik]

Oh dear. Sorry for all these posts.

Although that etch looked perfectly fine (really, really perfect) - there was a broken ground trace.
When first prototyping I will definitely check all traces with my DMM next time.

So - thank you all for your participation.

As promised, here is a complete build document. I am really not looking forward to boxing it up.

https://www.dropbox.com/s/kyueyu7cdfocvad/ZenBlender.pdf?dl=0

Cheers, Lars

[rsalinger]

Hi, lars-music!

Did you get around to boxing this up? Did you have any phase issues with the blend circuit? How did it sound? This looked quite interesting, at the time and it peaked my interest. I stumbled back upon this again, recently and I just might give it a go. Has this evolved, beyond the last circuit that you posted?

Rob

[roseblood11]

It sounds quite good as long as the gain doesn't get too high. I tried it with a tube screamer circuit at max gain or even with hard clipping (LED's to gnd), and didn't really like it. It's more like two sounds in parallel, they don't really blend together, maybe because the overdrive circuit affects attack and sustain of the signal too much. But it's really nice to make a mild overdrive more natural.
My version used an opamp as summing amp (and output buffer) and individual volume pots for the clean and overdriven signals. This adds a little more flexibility.
I never built this exact version, but it's the only picture with included schematic:
https://musikding.rocks/userImages/a6/347-a601b31f.gif
But I built versions 1 and 3 with success, which are in my musikding gallery as well. So, those layouts are verified, version 1.5 is NOT.

[lars-musik]

Hi Rob,
I built one of these according to the build document above. It worked good, no phasing problems (the blender isn't inverting the phase and nor is the zendrive).
But I don't really like the Zendrive and realized on top of that that I have no need for a clean blend on guitar pedals. So if you are asking for evolution - in my pedalboard ecosystem this one had a rather darwinian fate.

If you want to give it a try consider the build document verified.

Cheers, Lars