will a TS9 blow with 12V, 15V 18V??

[richon]

the OCD can handle 9 or even 18V.

can the TS9 (TS808) do the same, or 12 and 15V?

[Zben3129]

Is it a diy TS?

[Roobin]

Check the cap ratings inside. They will pretty much determine the max voltage. However, you should allow for a safety margin.

[richon]

its a DIY TS (i build it)

al caps are 50V or more..

I know the JRC4558D handles 15V or more...  but would something else burn or fail?

[Zben3129]

If you used 1/4 or greater wattage resistors (normal size ones), all your caps are 50 volts or more, and your IC is rated for 15v or more, than the only other thing i would suggest checking would be the transistors on the input and output buffers (JFETS iirc, been awhile). If those check out than I would imagine 15v should be fine.

P.S. - Does it have an led? If it does, you might wanna check the resistor leading to the led aswell to make sure the led can handle 15v


Zach

[kurtlives]

As long as the caps and IC can handle the extra voltage your set.

[Steben]

Is this intended to be used with higher treshold clipping?

[Mark Hammer]

Note that while the audio path can generally "take" higher supply voltages in a great many pedals, that is separate from a) whether the sound is improved in some measurable way, and b) whether other aspects of the pedal are amenable to that same supply voltage.  A great many of us use pedals whose tone is largely dependent on NOT having headroom.  In those instances, a higher supply voltage is not necessarily an improvement.  On the other hand, any effect whose objective is to remain clean while doing its thing MAY benefit from a higher supply voltage (though this is not a given).

For the nonaudio path, Boss uses an all-discrete flip-flop circuit to switch the FETs used for bypassing.  DOD also uses FETs for bypassing, but uses a 4007 CMOS chip for the flipping and flopping.  Where the discrete flip-flop of the Boss is perfectly content with 15-18V, CMOS chips get anxious as you approach 15v and dead-ish as you pass 15v.

The advice offered by Roobin to check the cap ratings is sound.  However, note that over the years, as cap size and cost have changed, pedal manufacturers tend to have shifted from using the lowest voltage rating they could get away with to simply buying caps in large volume that all had a "safe-under-any-circumstances" rating.  So, where 25-30 years ago a pedal-maker might have used "only" a 6.3V rated cap for the Vref circuit, since it was above the expected 4.5V and was both smaller and cheaper than a 16V rated cap, these days you are more than likely to see 35 and even 50V ratings no matter where the cap is in the circuit.  That is the long way of saying that the older the pedal, the greater the need to check the cap voltage ratings.  I suppose there are bound to be exceptions, but my gut sense is that if the pedal was made after 1990, you probably don't need to check the cap voltage ratings.  CMOS chips, on the other hand, have always been, and will remain, fry-able at >15V.

[R.G.]

the OCD can handle 9 or even 18V.

can the TS9 (TS808) do the same, or 12 and 15V?

WHY??

[MikeH]

the OCD can handle 9 or even 18V.

can the TS9 (TS808) do the same, or 12 and 15V?

WHY??

I suppose if you used it as more of a boost with the gain turned all of the way down it could benefit from more headroom; although,  I'm not sure if a TS will get more headroom from a higher input voltage.  Unless you remove the clipping diodes, that is.

[Mark Hammer]

It won't actually.

Consider what the typical input signal to a pedal is.  Now divide the space between that and the maximum voltage swing from any chips in there by the input signal.  So, if you could get a voltage swing of , say, 8v p-p and your input signal averaged 100mv p-p, you could squeeze a clean gain of 80x out of the pedal....in theory.  Of course, all you have to do is hit a chord, or even just play the two-string riff from Smoke on the Water with authority, and you are well over 100mv and into 250mv territory, where you're looking at a max clean (i.e., no headroom-related clipping) gain of 32x in our hypothetical scenario.

Well, all of that is great...IF there are no diodes in the picture.  Even if you set the gain to absolute minimum in a TS (which is 11.8x by my calculations) , those diodes start conducting at around 500-600mv and you would have to pick VERY gently for a gain of almost 12 applied to a typical guitar signal to not produce any clipping whatsoever.  I'm certainly not saying the resulting sound would be unpleasant - it may very well be exactly what you've been searching for all these years - but the odds of it being "clean" are slim to none.

Note that even if the voltage swing of the chip can be increased to , say 13v, by using a 15v supply, the diodes still impose a limit on signal amplitude which supercedes whatever potential is afforded by the change in supply voltage.  Certainly, if the user switched to a 15v supply for a TS, AND replaced the back-to-back 1N914 pair with a pair of LEDs or, heaven forbid, a 2+2 complement of LEDs to get a clipping threshold around 3v (i.e., things stay relatively clean until the signal reaches 6v p-p), then the higher supply voltage might afford more usable headroom, and the dream of having more dynamic range would remain alive.  However, the lesson to be learned is that the headroom does not always arise out of a mere change in supply voltage.  You always have to ask yourself whether there are other constraints to headroom.

[Steben]

I'ld say raising the supply voltage is a very nice action, when using discretes in overdrive, especially FET's.
Or one could use simple opamps as clean preamp. yet, I wonder why one should need clean boost higher than 3V p-t-p signal, which is easily obtained with a 9V battery. (this usually gets clean preamptubes into saturation).

[Mark Hammer]

That is an entirely reasonable and thoughtful suggestion.  One has to make a distinction between the intent of the pedal (distortion, EQ, compression, etc.) and its design or "mechanism" for achieving the sonic outcome.  In the case of diode-based clipping, the properties of the diode "mechanism" can supercede those of the power supply.  In the case of things like FET-based overdrives, where the clipping is not an explicit result of a pair of diodes, alterations to the power supply may well be very useful in extracting different characters out of the same circuit, without having to change anything other than perhaps biasing here and there.

As is currently being discussed in the ROD-10 thread, even Boss has taken similar FET-based discrete overdrive circuits and run them at different supply voltages to derive different personalities.

[MikeH]

It won't actually.

Consider what the typical input signal to a pedal is.  Now divide the space between that and the maximum voltage swing from any chips in there by the input signal.  So, if you could get a voltage swing of , say, 8v p-p and your input signal averaged 100mv p-p, you could squeeze a clean gain of 80x out of the pedal....in theory.  Of course, all you have to do is hit a chord, or even just play the two-string riff from Smoke on the Water with authority, and you are well over 100mv and into 250mv territory, where you're looking at a max clean (i.e., no headroom-related clipping) gain of 32x in our hypothetical scenario.

Well, all of that is great...IF there are no diodes in the picture.  Even if you set the gain to absolute minimum in a TS (which is 11.8x by my calculations) , those diodes start conducting at around 500-600mv and you would have to pick VERY gently for a gain of almost 12 applied to a typical guitar signal to not produce any clipping whatsoever.  I'm certainly not saying the resulting sound would be unpleasant - it may very well be exactly what you've been searching for all these years - but the odds of it being "clean" are slim to none.

Note that even if the voltage swing of the chip can be increased to , say 13v, by using a 15v supply, the diodes still impose a limit on signal amplitude which supercedes whatever potential is afforded by the change in supply voltage.  Certainly, if the user switched to a 15v supply for a TS, AND replaced the back-to-back 1N914 pair with a pair of LEDs or, heaven forbid, a 2+2 complement of LEDs to get a clipping threshold around 3v (i.e., things stay relatively clean until the signal reaches 6v p-p), then the higher supply voltage might afford more usable headroom, and the dream of having more dynamic range would remain alive.  However, the lesson to be learned is that the headroom does not always arise out of a mere change in supply voltage.  You always have to ask yourself whether there are other constraints to headroom.

As I suspected... Hey!  I think I've learned something!

[JDoyle]

It may also be important to note that the TS, though often thought to be otherwise, is CROSSOVER DISTORTION.

When the diodes clip, the gain reduces to one and the actual input signal then comes through. Taking a diode voltage of 0.6V, a 100mV input signal, and the gain of the circuit set to 100, the portion of the input signal that gets the full gain is that which is .006V or lower, after that the diodes conduct and the actual input signal is fairly accurately reproduced. This is simplified, but it's how it works - the middle portion of the signal about 'ground' multiplied by the gain is the distorted portion of the signal - the TS doesn't 'clip' so much as 'distort'.

So, if you are nailing the input of the TS with a boosted signal, the higher headroom of a higher voltage may be warranted - though the larger the input signal, the less the distortion created in the TS matters in terms of it's overall percentage of the output. Not to mention, the less you may LIKE the distortion of the TS with a large input signal - I think it ends up pretty buzzy...

Regards,

Jay Doyle

[Ben N]

Dave Barber advertises that his Direct Drive (a somewhat TS-ish pedal, although Dave bristles at the use of the C-word) will work fine on 18v, so I tried it for a while on Mike's theory. At first I thought I heard something like additional clarity, then eventually concluded that that was mostly my imagination, and not worth the extra wall-wart in the rig. I do appreciate that Dave's build quality is so good, and that he he understands that his customers will want to try stuff (which is why he also sockets his ICs and posts mods on his website).

[richon]

all my thoughts had been answer 

i just was looking for a reason no to experiment this... i was thinking about the CLIPPING Voltage of different diodos, leds,  and so on....

[Steben]

It may also be important to note that the TS, though often thought to be otherwise, is CROSSOVER DISTORTION.

When the diodes clip, the gain reduces to one and the actual input signal then comes through. Taking a diode voltage of 0.6V, a 100mV input signal, and the gain of the circuit set to 100, the portion of the input signal that gets the full gain is that which is .006V or lower, after that the diodes conduct and the actual input signal is fairly accurately reproduced. This is simplified, but it's how it works - the middle portion of the signal about 'ground' multiplied by the gain is the distorted portion of the signal - the TS doesn't 'clip' so much as 'distort'.

So, if you are nailing the input of the TS with a boosted signal, the higher headroom of a higher voltage may be warranted - though the larger the input signal, the less the distortion created in the TS matters in terms of it's overall percentage of the output. Not to mention, the less you may LIKE the distortion of the TS with a large input signal - I think it ends up pretty buzzy...

Regards,

Jay Doyle

That's a bit what I was getting too as well.
It is, however, a different cross-over than that in Class B amps. I mean, with the TS there is a steep cross-over between + and -, in class B amps, it is flat. That gives a different shade.

It is true that the larger the input, the larger the treshold of the clipping should be to obtain the same effect.
A fair humbucker into a GE-diode pair tube screamer will sound BAD. The signal is way bigger than the clipping treshold.
It all comes down to scaling, if you intend to have a booster in front most of the time, you could better stick LEDs in or MOSfets.

Another trick which originates from the same mechanism is the double TS in series. Here the first distorted signal becomes the dry signal of the second TS. As you may know now, the signal gets kinda added to the 0.6v treshold. If you add a soft square wave on top af a hard one, you get in fact at high gain the same effect as a double treshold but with double gain.

[wampcat1]

Dave Barber advertises that his Direct Drive (a somewhat TS-ish pedal, although Dave bristles at the use of the C-word) will work fine on 18v, so I tried it for a while on Mike's theory. At first I thought I heard something like additional clarity, then eventually concluded that that was mostly my imagination, and not worth the extra wall-wart in the rig. I do appreciate that Dave's build quality is so good, and that he he understands that his customers will want to try stuff (which is why he also sockets his ICs and posts mods on his website).

Just to clarify for others reading... the direct drive isn't a clone of the ts at all. It's actually quite different, other than the fact the first stage is similar in function. There is also diodes to ground after that stage, a LP filter for a high control and the second stage is more of a bass boosting stage. It sounds alot fatter and flatter eq wise (IMO).

Not picking on you, just wanted to clarify so no one on another forum would say "I read on diystompboxes that it is a clone!!"

bw

[Ben N]

Thanks for the clarification, Brian. I didn't mean to suggest that the circuit is a clone--I haven't seen a schematic, I haven't reversed mine, and I know only what Dave has said, that the bass-restoring eq at the end is based on a RIAA-standard phono preamp eq. All I meant is that it is based on a dual opamp, and functionally it occupies pretty much the same sonic territory, that is, a midrange-emphasized medium overdrive. I prefer it to a TS-9, say, because it is true bypass, because the mid-hump is somewhat less pronounced and wider, which allows it to do a fair Marshall simulation with the tone control pulled up,  because I find the control ranges very useful, and because it takes boosting well.