Chuck D. Bones
Circuit Wizard
At long last, it's done. Well, except for the paint job. Those are two modd'ed Photon Vibe boards inside. Left channel is yellow, Right is blue. Controls are (top row) BLEND, COLOR, BLEND, COLOR (bottom row) SPEED, DEPTH, SPEED, DEPTH. Top switch is SYNC, bottom switch is Series. The two channels can operate completely independently, they can be sync'ed and they can be run in series. Yellow and blue LEDs are the THROB indicators; green is ENGAGE.
Here's the Robot Porn. Look carefully for mods. Most are on the top of the boards. Note that all resistors are 1/8W. The Photon is a tight board. Film caps in the audio path are WIMA or Panasonic. The one electrolytic cap in the audio path (C21) is tantalum. LFO caps are tantalum for minimum leakage.
With the light shields installed. They were cut from a 6-pack of Abita Purple Haze lager. They butt up against the bottom cover.
The LDRs are duals. I used them because they have a huge range, are matched, and I had 'em. The LDRs called out in the BOM would have worked just fine.
I made a few mods to the LFO and lamp driver, including the addition of a SYNC switch that allows the Left Channel LFO to influence the Right Channel LFO. I also relocated the THROB LEDs so that they are brighter and are not dependent on the DEPTH setting. I ditched the HIGH, LOW and LEVEL controls. I hardwired the gain for unity. One of the LED trimmers was relocated to the front panel as the COLOR (COLOUR for you Aussies & Kiwis) control. The BLEND trimmer was rewired and moved to the front panel. At max, the blend is 50/50. At min, the blend is 20% dry and 80% wet.
Sounds great! By tweaking the SPEED controls in Sync mode the LFOs can be made to run in phase, out of phase or chaotic. Something I didn't expect (but should have) was the effect of the channels mixing in the room. You don't just get the swirling from left to right, the phase between the speakers keeps shifting and the effect is the blend is constantly moving. It's glorious.
There is a minor issue, which I'm still working on, in that if the lamp is allowed to turn completely off, there is a mild tick. Proper setting of the DEPTH and COLOR controls prevents the tick, but I'm looking to find another solution. NB: the LFO is not ticking, it is the lamp driver interacting with the power supply that causes the tick.
The Photon Vibe is a pretty good circuit, but it has one major problem in the power supply: The 15V regulator doesn't regulate. The reason is inadequate headroom. 78Lxx regulators need at least 2V of headroom under the best conditions, preferably more. The charge pump makes just over 17V when the lamp is off. When the lamp is on, the voltage drops to 16.5V or lower. The "18V" rail bounces around at the LFO rate and consequently the "15V" rail and Vref do the same. You can put a scope on the output of the pedal and see the LFO signal bleeding thru. My solution is to replace the 78L15 with a 78L12. Now the opamps get clean power and a stable Vref.
LFO mods
The LFO's bias is derived from the "18V" supply, which as I said, bounces around. Better to bias the LFO from something stable, so I use the regulated 12V rail to power the bias circuit. R11 now connects to 12V instead of 18V and R8 was changed to 10K.
At the bottom of the SPEED range, the LFO signal gets very weak. This is a consequence of the LFO design. To counter that, I did two things: increased C8 to 2.2uF and added 270K resistors in parallel with the two halves of the SPEED pot. The SPEED range is now 0.65Hz to 4Hz. The LFO amplitude still drops at lower speeds, but not nearly as much.
I replaced Q1 and Q2 with an MPSA13. That was a no-brainer and freed up some pads for wiring in the THROB LED between the collector and the 18V rail. Notice that I mounted Q1 "backwards" to take advantage of the pads freed up by removing Q2.
A SYNC switch was added to connect the output of the Left LFO (pin 2 of the SPEED pot) to the junction of C7 & C8 on the right LFO via a 100nF cap. Look carefully at the pix, that cap is an MLCC soldered directly to the SYNC switch. I tried several different configurations for Sync and this worked the best across the Speed range. With Sync engaged, setting the LFO SPEEDs the same (or very nearly the same) causes the Left & Right LFOs to sync up in phase. Move one of the SPEED controls up or down a little and they stay locked, but their relative phase changes. It's not hard to get them close to 180 degrees out of phase. Move the SPEED controls further apart and the right channel LFO locks onto some fraction of the left LFO. I've been able to achieve simple fractions like 1/2 or 2/3, as well as crazy fractions like 11/15. At some settings, the right channel LFO gets pretty chaotic.
Lamp Driver mods
I replaced Q3 with an MPSA13, which raises the driver's input impedance and allows me to change TR3, R22, R24 and R26 so that COLOR and DEPTH don't interact. R22 is now 100K, R24 is 10K, R26 is 68K and TR3 is replaced by a B250K pot. R32 is deleted because the LED is now wired to the LFO instead of the lamp driver.
Audio Path mods
I left the phase shifter stages alone, they are fine as-is. The LOW pot and C2 are deleted. C1 is 22nF. R33 and R4 are 2.2M. R5 is 75K. The HIGH pot, R25 and C18 are deleted. TR1 is hardwired. R16 is replaced with a C10K BLEND pot, a 6.8K resistor and a 10uF cap (see schematic for clarification). R28 is deleted and R30 is a jumper. The LEVEL pot is deleted, pads 2 and 3 are jumpered and a 100K anti-pop resistor is connect from pad 1 to pad 2.
A SERIES switch connects the Left channel input to either the input jack of the Right channel output.
That's about it. This box is definitely getting an acrylic pour. The schematic is on four pages for clarity, PDF is attached. Note that I have not yet installed the Disable. It kills the lamp drive when the pedal is in bypass mode. The intent is to reduce power consumption in bypass mode and prolong lamp life.
Here's the Robot Porn. Look carefully for mods. Most are on the top of the boards. Note that all resistors are 1/8W. The Photon is a tight board. Film caps in the audio path are WIMA or Panasonic. The one electrolytic cap in the audio path (C21) is tantalum. LFO caps are tantalum for minimum leakage.
With the light shields installed. They were cut from a 6-pack of Abita Purple Haze lager. They butt up against the bottom cover.
The LDRs are duals. I used them because they have a huge range, are matched, and I had 'em. The LDRs called out in the BOM would have worked just fine.
I made a few mods to the LFO and lamp driver, including the addition of a SYNC switch that allows the Left Channel LFO to influence the Right Channel LFO. I also relocated the THROB LEDs so that they are brighter and are not dependent on the DEPTH setting. I ditched the HIGH, LOW and LEVEL controls. I hardwired the gain for unity. One of the LED trimmers was relocated to the front panel as the COLOR (COLOUR for you Aussies & Kiwis) control. The BLEND trimmer was rewired and moved to the front panel. At max, the blend is 50/50. At min, the blend is 20% dry and 80% wet.
Sounds great! By tweaking the SPEED controls in Sync mode the LFOs can be made to run in phase, out of phase or chaotic. Something I didn't expect (but should have) was the effect of the channels mixing in the room. You don't just get the swirling from left to right, the phase between the speakers keeps shifting and the effect is the blend is constantly moving. It's glorious.
There is a minor issue, which I'm still working on, in that if the lamp is allowed to turn completely off, there is a mild tick. Proper setting of the DEPTH and COLOR controls prevents the tick, but I'm looking to find another solution. NB: the LFO is not ticking, it is the lamp driver interacting with the power supply that causes the tick.
The Photon Vibe is a pretty good circuit, but it has one major problem in the power supply: The 15V regulator doesn't regulate. The reason is inadequate headroom. 78Lxx regulators need at least 2V of headroom under the best conditions, preferably more. The charge pump makes just over 17V when the lamp is off. When the lamp is on, the voltage drops to 16.5V or lower. The "18V" rail bounces around at the LFO rate and consequently the "15V" rail and Vref do the same. You can put a scope on the output of the pedal and see the LFO signal bleeding thru. My solution is to replace the 78L15 with a 78L12. Now the opamps get clean power and a stable Vref.
LFO mods
The LFO's bias is derived from the "18V" supply, which as I said, bounces around. Better to bias the LFO from something stable, so I use the regulated 12V rail to power the bias circuit. R11 now connects to 12V instead of 18V and R8 was changed to 10K.
At the bottom of the SPEED range, the LFO signal gets very weak. This is a consequence of the LFO design. To counter that, I did two things: increased C8 to 2.2uF and added 270K resistors in parallel with the two halves of the SPEED pot. The SPEED range is now 0.65Hz to 4Hz. The LFO amplitude still drops at lower speeds, but not nearly as much.
I replaced Q1 and Q2 with an MPSA13. That was a no-brainer and freed up some pads for wiring in the THROB LED between the collector and the 18V rail. Notice that I mounted Q1 "backwards" to take advantage of the pads freed up by removing Q2.
A SYNC switch was added to connect the output of the Left LFO (pin 2 of the SPEED pot) to the junction of C7 & C8 on the right LFO via a 100nF cap. Look carefully at the pix, that cap is an MLCC soldered directly to the SYNC switch. I tried several different configurations for Sync and this worked the best across the Speed range. With Sync engaged, setting the LFO SPEEDs the same (or very nearly the same) causes the Left & Right LFOs to sync up in phase. Move one of the SPEED controls up or down a little and they stay locked, but their relative phase changes. It's not hard to get them close to 180 degrees out of phase. Move the SPEED controls further apart and the right channel LFO locks onto some fraction of the left LFO. I've been able to achieve simple fractions like 1/2 or 2/3, as well as crazy fractions like 11/15. At some settings, the right channel LFO gets pretty chaotic.
Lamp Driver mods
I replaced Q3 with an MPSA13, which raises the driver's input impedance and allows me to change TR3, R22, R24 and R26 so that COLOR and DEPTH don't interact. R22 is now 100K, R24 is 10K, R26 is 68K and TR3 is replaced by a B250K pot. R32 is deleted because the LED is now wired to the LFO instead of the lamp driver.
Audio Path mods
I left the phase shifter stages alone, they are fine as-is. The LOW pot and C2 are deleted. C1 is 22nF. R33 and R4 are 2.2M. R5 is 75K. The HIGH pot, R25 and C18 are deleted. TR1 is hardwired. R16 is replaced with a C10K BLEND pot, a 6.8K resistor and a 10uF cap (see schematic for clarification). R28 is deleted and R30 is a jumper. The LEVEL pot is deleted, pads 2 and 3 are jumpered and a 100K anti-pop resistor is connect from pad 1 to pad 2.
A SERIES switch connects the Left channel input to either the input jack of the Right channel output.
That's about it. This box is definitely getting an acrylic pour. The schematic is on four pages for clarity, PDF is attached. Note that I have not yet installed the Disable. It kills the lamp drive when the pedal is in bypass mode. The intent is to reduce power consumption in bypass mode and prolong lamp life.
Last edited:
