TTSH V2 mods, build issues (Mostly what I have learned by reading Muff Wiggler posts) R Grieb Last update: 7/28/2016 From Muff Wiggler/build notes: 1) VCO 680 pf timing caps should be better quality than simple ceramic types. Either polystyrene, mica, or polypropylene would be a better choice for stability. 2) Change to flashing LED circuit base drive to get complete switching 3) Possible insertion of resistor in series with collector of current source in flashing LED string to reduce transistor dissipation/heating. I used two 330 ohm 1/2W resistors in series. Transistor dissipation depends on how bright you like your LEDs 4) Possible change to 3.3K resistor in LED brightness adjust circuit. I changed it to 1.3K to allow the LED's to be turned off with the trimmer, if desired. Probably 1K or just a jumper would be a better choice as some LEDs still glow a little with my value. 5) Nordcore's mod to reduce Johnson noise in reverb circuit 6) Nordcore's mod to AR attack circuit, enabling faster attacks, and longer times. 7) Reverb tank needs to be kept away from any sort of power transformer, and shielded cable should be used to connect it to the main pcb 8) V2 PCB has places for four non-polarized caps in series with the inputs to the VCF. This should remove DC bias and reduce thumping when the VCA opens up. You will need to cut the traces to install these components. I used 1 uF metallized polyester caps. 9) Seems like speaker outputs should feed through each speaker, then to the headphone jack, where they connect to ground if headphones are not installed. If headphones are installed, they will connect to ground through the headphones, so the speakers will still be on, but at reduced volume. This is better than wiring to the headphone jack first and then to the speakers. In this case, if a mono plug is inserted into the headphone jack, one speaker amplifier output will be shorted to ground. This could easily fry the 10 ohm amplifier emitter resistors, since they are only 1/4W. 10) Several people have noted the difference between the AR and ADSR maximum level. Not sure if this also applies to the 2600, or what mods are indicated, if any. Other things I have noticed: 11) Seems to me that Ring Mod Pos Null and Neg Null trimpot silkscreen labels are swapped, based on comparison with the 2600 and how the calibration went. 12) ARP 2600 uses 1/2 W 100 ohm resistor in series with the input to the reverb tank. The input impedance of the tank is just a few ohms, so if we have 5V RMS at the drive output, that's 1/4W that the resistor would have to dissipate, which is too much for a 1/4W resistor. 13) If Murata DC-DC converters are installed, and their output is 15V, that's not enough voltage at the input of LDO's to create the regulated 15V output. (I later discovered that when using the DC-DC converter with +/-15V outputs, the LDO outputs were supposed to be set for 14.6V, which would be fine. I did not use the DC-DC converter.)