Electro Harmonix Bad Stone

The original Bad Stone Phase Shifter sports my favorite Electro Harmonix graphic ever. A great conversation starter. It is unknown what that graphic is, but "Mummy" is the usual description.

Bad Stone General Info

The Bad Stone began as a matched JFET based phaser, probably inspired by the Maestro PS-1. The pedal was designed by Bob Bednarz. Release date is murky, but it is shown in 1974 advertisements. It is notable for including feedback via its "Blend" control, (later versions changed the name to "Feedback") which the earlier Maestro PS-1 and MXR Phase 90 units lacked.

The original graphics feature a grotesque face. This is usually called the "Mummy" graphic, or V1. Some early oddities may exist. In some EH advertisements, there is a "Ram's Head" Bad Stone shown. I've never seen a photo of an actual version produced this way. Effects Database.com has a photo of one with a full face rubber pad, covering up where the Mummy would be.

The problem with matched JFET phasers is that the factory has to match sets of JFETs manually, and this costs time and money. Around the same time this pedal was coming out, David Cockerell came to work at Electro Harmonix and he brought with him an OTA phaser design he had already used in the EMS Synthi Hi Fli in 1972. The OTA chips required no matching. This appealed to EH owner Mike Matthews, and he put Cockerell's design into production right on the heels of the Bad Stone. The OTA phaser design became the Small Stone, and it also included feedback via the "Color" switch.

The discrete JFET Bad Stone was replaced by a CD4009 MOSFET version. Using the CD4009 chip meant no more matching while retaining the basic FET topology.

The most common type of Bad Stone seen has three controls for rate, feedback, and manual shift. This usually has orange and black silkscreen, and the graphic design is similar to the other late '70s EH product line. These are called "V2" sometimes. This version is CD4009 based.

A rare wah type version is out there. The gut shots I've seen have the CD4009 and a CA3246 BJT array. I haven't been able to catch the board designator yet for this model. EH Man's product list implies it should be EH 5801, but the gut shots appear to be showing something else. We'll just have to wait until I get one to come through the shop.

Official JFET based Bad Stone schematic. This appears to refer to a "wah" enclosure type pedal (note "pedal pot" has a "toe down" marking). The Mummy version is slightly different. Phase shift stages have the same basic layout. See my traced schematic below.

Early version partial schematic from a traced EH 5800 unit. Note output buffer that isn't present on the official schematic, and how the dry kill is labeled "Color" and the feedback control is labeled "Blend." The unmarked Blend is a 1M pot, as it is on the official schematic.

Discrete JFET Version Info

There are at least two versions of the discrete JFET version. What I presume to be the older one uses two triple op amps with no markings on the package (see more info on this below). The later version switched to common quad op amps packages. Circuit board layout was changed between the two versions.

The early versions have a lot of flying components, cut traces, and do not match any official schematic I have seen. This may have been the result of EH tweaking the design along the way.

Original advertisement description:

"This PHASE SHIFTER cycles any instrument or microphone signal through 720 degrees of phase rotation, generating a bubbling water-like doppler phase shift. Comparable in quality of sound to a studio phase shifter. Comes with Blend and Rate Control knobs that allows adjustment for depth of effect and changes rate at which effect occurs. There, also, is a special toggle switch which produces standard swoosh phase shifted mode or frequency-warble mode."

In the boxes I've seen, the "Blend" is a manual feedback control, and the "Color" toggle switch disconnects the dry signal. The "Color" switch does produce pitch shifting vibrato, provided the rate is around a few Hz (not too slow, not too fast), and the "Blend" (feedback) isn't too high (best pitch bend when "Blend" is at minimum).

Discrete JFET Calibration

There should be a FET bias trim pot and a LFO depth trim pot. This is the same way a Maestro PS-1 is calibrated. Turn the LFO depth to minimum (no LFO modulation), then adjust the FET bias pot to the mid point of the sweep. (I find it easiest to use a swept function generator and observe the phase notches moving on an oscilloscope.) Raise the LFO amount until you have maximum symmetrical sweep.

There is also a trim pot that adjusts the resistance of the wet signal where it sums with the dry signal. I'm not sure what the original intent of this was, other than balancing the two signals.

I haven't seen enough Mummy versions to know if all 3 of these trims are usually present or not. One example only had the FET bias trim, but appeared to have been modded at some point.

Mystery Triple Op Amp Pinout

The early JFET based "Mummy" Bad Stones use two triple op amp ICs for the six phase stages. All of the markings were scratched off of these pins, so the part number is unknown. These are easy to identify as they are the only 14 pin DIP packages on the board.

Here is the pinout:

Pin 01: Bias

Pin 02: OP1 output

Pin 03: OP2 -in

Pin 04: OP2 +in

Pin 05: OP3 +in

Pin 06: OP3 -in

Pin 07: nc

Pin 08: nc

Pin 09: OP3 output

Pin 10: V-

Pin 11: OP2 output

Pin 12: OP1 +in

Pin 13: OP1 -in

Pin 14: V+

The following photos are from a unit that came to me with a replaced triple op amp. A previous tech had figured out a way to turn two SIP dual op amps into a suitable replacement.

EH 5800

Close up of the dual SIP being used as a replacement for a rare triple op amp. I did not make this. This was inside when I opened up the pedal.

At this angle you can read M5218L on the SIP package.

The solder side. The trim pot looks like it was replaced, probably during the triple op amp repair. After servicing another Mummy with 3 trim pots, I think this one had some trims removed.

That says EH 5800. I think I might have even sucked some of the solder away to confirm that after I took this photo. EH 5850 is the same circuit but with a layout for the much more common LM324 quad op amp.

I think I was trying to capture that this 2N3563 transistor only has 2 legs... and it's supposed to be that way. There are 2 Bad Stone schematics. This matches the one with discrete FETs and you will see the transistor-as-diode in that schematic.