- Build Rating
Hello all! First build report here.
This is a PedalPCB Maleficent Mids built into a Gorva C65 White Matte enclosure. Red LED, mixed knobs, for a ghostly-mean-cyborg kind of look!
I started building my own DIY pedals in 2023 and I actually built this in 2024. Since then I went through more builds and I learned a ton reading these forums, both theory and practical tips.
I just revamped this with nicer wiring, an overall cleanup and some mods.
The Maleficent Mids is a clone of the JHS Haunting Mids. Overall this is a nice semi-parametric EQ for tweaking mids. I rocked this on my pedalboard for a while before moving to an HX Stomp for live shows, but I'm still enjoying it on my analog pedalboard at home.
I did have some itches with this circuit though:
1) Unity VOL was somewhere around 10:00 and hard to dial in due to being linear (B10K). Here's what I did to get unity at noon. I took an A10K pot I had and measured the actual resistance between lugs 1 and 2 with pot at noon. In theory we get 1k for a log pot but this varies -- I actually measured 1k5. So my A10K volume pot would have gain 1k5/10k = 0.15 at noon. Now, the input stage at IC1.1 has gain 1 + 10k/4k7 = 3.1 and the filter stage around IC2 seemingly had gain 1 with MIDS at noon (found this by simulating with Falstad afilter). The overall gain with VOL at noon is then 3.1 x 1 x [output stage gain] x 0.15 and I want this to be 1. So gain of the IC1.2 output stage must be 1/(3.1 x 0.15) which is 2.15. But IC1.2 is an inverting amplifier and its gain is R17/R16. So R17 = 2.15 x 4k7 = 10.1k -- so I replaced R17 with 10K instead of 4k7. Note that if your A10K measures a proper 1k between lugs 1 and 2, you'd end up with having to use R17=15k. I put R17 on sockets in case I ever need to tweak it again...
2) As others noticed, this circuit has pretty low input headroom. The active humbucker of my Schecter SVSS makes it clip! I did many attempts to improve this, but to no avail. Attempts include 1) swapping IC1 and 2, i.e. putting the NE5532 in the filter stage rather than as input/output stages; 2) The 4u7 input cap was unnecessarily large (input HPF at 0.7 Hz), so I dropped it to 47n (input HPF at 7 Hz) in case the sub-bass could cause clipping. Not much luck. One last thing I haven't tried yet is reducing R3 from 10k to 4k7 so that the input signal is only boosted 2x instead of 3x before going through the filter. Yet another would be running it at 18V... Anyways. This effect does not appear when cutting mids rather than boosting, though. So I can sort of boost lower mids by cutting high-mids/treble and cranking the VOLUME. Some interesting tones to be found that way. The clipping could also be considered a feature, since it can give a rudimentary overdriven mid boost flavor when really pushing the frontend.
3) The frequency range for SWEEP is a bit too high for my liking (380 Hz to 7.3kHz with 750 Hz at noon). The range is controlled by C3, C4, R9 and R10. I figured it can be computed like so: fMax = 1/(2 x pi x sqrt(R9 x R10 x C3 x C4)). Add +50K and +100K to R9 and R10 in the formula to get fNoon and fMin respectively. Replacing C4 with 3n3 brings (260 Hz, 600 Hz, 5kHz) which would seem nicer to me. Sadly C4 is stuck under the high B100K dual pot, so I could not replace it.
4) The Q is too wide for my liking. I read it was estimated at 0.5 on LO and 1.5 on HI. This can't really bring the parked wah sound which I was keen for... I would have preferred something like LO Q=1.5 and HI Q=3-5. I did a LOT of tinkering with the filter using Falstad afilter. I could not find a mod that would bring tighter Q without having to replace the whole set of components around IC2.1. So I left it this way... If I were to build this again I would try the values from dylan159's Core Shaper: sweeps from 300 Hz to 3kHz and with Q = 1 or 2.4.
Of all these changes / experimentations I really do recommend 1) aka changing VOL to a log pot and tweaking R17 for unity at noon. It's a great quality of life improvement for using the pedal. The R17 tweak increases the total available volume boost as well.
Some other changes:
- Dropped C102 as someone else suggested, since it has no effect.
- Put R100 on sockets to make changing the LED easier. I initially used a violet LED. Changed it to red so had to change the CLR.
I'm pretty happy with how it is now! Boosting with SWEEP at noon (750 Hz) and pushing its input to bring it into overdrive can get those Money For Nothing tones. Setting SWEEP high to boost high mids can add bite to a following drive pedal. Scooping the mids before or after an effect can sound interesting too. So many uses!
Overall, 4 stars on the stock JHS circuit as it's hard to mod. But I'd give 5 stars to the PedalPCB PCB! Easy to build. I'd love those GND pads to be wider though. They're tied to the ground plane too, so removing / sucking solder from them while revamping the wiring was a bit of a pain (I suppose my iron wasn't hot enough).
Edit: forgot the pics.
This is a PedalPCB Maleficent Mids built into a Gorva C65 White Matte enclosure. Red LED, mixed knobs, for a ghostly-mean-cyborg kind of look!
I started building my own DIY pedals in 2023 and I actually built this in 2024. Since then I went through more builds and I learned a ton reading these forums, both theory and practical tips.
I just revamped this with nicer wiring, an overall cleanup and some mods.
The Maleficent Mids is a clone of the JHS Haunting Mids. Overall this is a nice semi-parametric EQ for tweaking mids. I rocked this on my pedalboard for a while before moving to an HX Stomp for live shows, but I'm still enjoying it on my analog pedalboard at home.
I did have some itches with this circuit though:
1) Unity VOL was somewhere around 10:00 and hard to dial in due to being linear (B10K). Here's what I did to get unity at noon. I took an A10K pot I had and measured the actual resistance between lugs 1 and 2 with pot at noon. In theory we get 1k for a log pot but this varies -- I actually measured 1k5. So my A10K volume pot would have gain 1k5/10k = 0.15 at noon. Now, the input stage at IC1.1 has gain 1 + 10k/4k7 = 3.1 and the filter stage around IC2 seemingly had gain 1 with MIDS at noon (found this by simulating with Falstad afilter). The overall gain with VOL at noon is then 3.1 x 1 x [output stage gain] x 0.15 and I want this to be 1. So gain of the IC1.2 output stage must be 1/(3.1 x 0.15) which is 2.15. But IC1.2 is an inverting amplifier and its gain is R17/R16. So R17 = 2.15 x 4k7 = 10.1k -- so I replaced R17 with 10K instead of 4k7. Note that if your A10K measures a proper 1k between lugs 1 and 2, you'd end up with having to use R17=15k. I put R17 on sockets in case I ever need to tweak it again...
2) As others noticed, this circuit has pretty low input headroom. The active humbucker of my Schecter SVSS makes it clip! I did many attempts to improve this, but to no avail. Attempts include 1) swapping IC1 and 2, i.e. putting the NE5532 in the filter stage rather than as input/output stages; 2) The 4u7 input cap was unnecessarily large (input HPF at 0.7 Hz), so I dropped it to 47n (input HPF at 7 Hz) in case the sub-bass could cause clipping. Not much luck. One last thing I haven't tried yet is reducing R3 from 10k to 4k7 so that the input signal is only boosted 2x instead of 3x before going through the filter. Yet another would be running it at 18V... Anyways. This effect does not appear when cutting mids rather than boosting, though. So I can sort of boost lower mids by cutting high-mids/treble and cranking the VOLUME. Some interesting tones to be found that way. The clipping could also be considered a feature, since it can give a rudimentary overdriven mid boost flavor when really pushing the frontend.
3) The frequency range for SWEEP is a bit too high for my liking (380 Hz to 7.3kHz with 750 Hz at noon). The range is controlled by C3, C4, R9 and R10. I figured it can be computed like so: fMax = 1/(2 x pi x sqrt(R9 x R10 x C3 x C4)). Add +50K and +100K to R9 and R10 in the formula to get fNoon and fMin respectively. Replacing C4 with 3n3 brings (260 Hz, 600 Hz, 5kHz) which would seem nicer to me. Sadly C4 is stuck under the high B100K dual pot, so I could not replace it.
4) The Q is too wide for my liking. I read it was estimated at 0.5 on LO and 1.5 on HI. This can't really bring the parked wah sound which I was keen for... I would have preferred something like LO Q=1.5 and HI Q=3-5. I did a LOT of tinkering with the filter using Falstad afilter. I could not find a mod that would bring tighter Q without having to replace the whole set of components around IC2.1. So I left it this way... If I were to build this again I would try the values from dylan159's Core Shaper: sweeps from 300 Hz to 3kHz and with Q = 1 or 2.4.
Of all these changes / experimentations I really do recommend 1) aka changing VOL to a log pot and tweaking R17 for unity at noon. It's a great quality of life improvement for using the pedal. The R17 tweak increases the total available volume boost as well.
Some other changes:
- Dropped C102 as someone else suggested, since it has no effect.
- Put R100 on sockets to make changing the LED easier. I initially used a violet LED. Changed it to red so had to change the CLR.
I'm pretty happy with how it is now! Boosting with SWEEP at noon (750 Hz) and pushing its input to bring it into overdrive can get those Money For Nothing tones. Setting SWEEP high to boost high mids can add bite to a following drive pedal. Scooping the mids before or after an effect can sound interesting too. So many uses!
Overall, 4 stars on the stock JHS circuit as it's hard to mod. But I'd give 5 stars to the PedalPCB PCB! Easy to build. I'd love those GND pads to be wider though. They're tied to the ground plane too, so removing / sucking solder from them while revamping the wiring was a bit of a pain (I suppose my iron wasn't hot enough).
Edit: forgot the pics.
Last edited: