Has anyone tried modding a Pro Co Rat to run on 18V supply?

[Jasonmatthew911]

I'm modding a vintage re-issue Rat, and I was wondering if anyone has modded a Rat to run on 18V, or if anyone has run a Rat at 18V with no problems and good results?....The only think I could think of is maybe adding an 18V Zener diode to limit the 18V, and maybe add a clip on heat sink since the LM308N operates on 18V max.....Can anyone help me with this?....Thanks.

[StereoKills]

Make sure all the caps are rated for at least 25v. I don't know about the Rat in particular, but some manufacturers tend to use 12-15v rated caps to save money/space.

[Jasonmatthew911]

Make sure all the caps are rated for at least 25v. I don't know about the Rat in particular, but some manufacturers tend to use 12-15v rated caps to save money/space.

Good idea...Thanks....I wonder why more people haven't experimented with the Rat at 18V, just to give it more juice to really push some more volume and have more headroom.

[PRR]

Pro Co Rat is a distortion unit.

If you raise the power voltage, you will have to work twice as hard to make it distort.

I'm missing the point.

BTW, LM308N can take 36V total; the "+/-18V" spec is for split-supply.

[Johan]

as PRR says..it's a distortionbox..the signal gets shunted to ground through two diodes...raising the supplyvoltage is a pointless exercise
J

[R.G.]

But on the other hand, if it has no effect at all, and does not actually kill devices or operate them in ways that make their lifetimes contract down to days, hours or minutes, it can be considered to be head and shoulders above many "mods". 

[Jasonmatthew911]

Actually I just finished my modding and had success with it...I have a Vintage Re-issue Rat modded by Robert Keeley with metal can LM308AH and another '91 Re-issue (LM308N) modded by me, I basically changed all the caps to metal film and silver mica, and raised the value of a few for more bass response, I also changed the 100uf/16V electrolytic cap to a 220uf/35V, since it was the only cap below 18V...I added a clip on heat sink to the chip  and an 18V 1N5355 zener diode just in case, and finally changed the power jack to a standard boss style to use with my 18V dunlop power supply......I compared it to my Robert keeley Rat on 9V, while my modded Rat was running on 18V, and it sounded louder, and distorts just as much, but with more headroom and low end than the Keeley Rat...The 18V makes it sound bigger and more alive in my opinion...

[R.G.]

Good. We'll call this a highly successful mod then. 

[Jasonmatthew911]

Yeah, the difference from 9V to 18V isn't a huge difference, but I do notice slightly more clarity and headroom, as well as a more amp-like tone with 18V, which is cool to take a little bit of the fuzziness out at higher gain settings, but it still feels pretty dirty like a rat, only slightly bigger with slightly more clarity...I think that making a Rat capable of running on 18V, plus changing some caps to metal film (Changing some values slightly),  and silver mica, is definitely an improvement...A metal can LM308AH, would probably improve it some more even....Maybe I'll try a Metal can if I get my hands on one.

[R.G.]

OK.
Here's the theoretical thing, though. See that diode clipping pair to ground? Those limit the output voltage. Going to 18V on the opamp may have a tiny change in how much voltage it can drive into them. Maybe. But not much. And even if it does, I wonder how much "clarity" and "headroom" it could add.

See, "headroom" usually means "how much more could I drive it without distorting".  In this circuit, the result is - none. The diodes start clipping at the same place, no matter how much further into clipping they're driven. Of course, you could be meaning something entirely different by the word "headroom", as I am reminded by the most memorable of William Jefferson Clinton's saying - "It all depends on what your definition of "is" is...".

I suspect that the changing of capacitors has much more to do with what you're hearing than the power supply change to 18V. Could be wrong, but that's my suspicion.

[Bill Mountain]

Wouldn't the increased voltage change the way the 308 clips before it hits the diodes?

[nexekho]

I do remember reading that the limited slew rate of the op amp used in a Rat is part of the sound - by doubling the voltage, does this mean that the op amp takes twice as long to swing through the full range?  Might this change the sound a little?

[anchovie]

Mustn't forget the psychological aspect of modding - if you want it to sound different, it will. 

[richon]

maybe he changed the input cap value (x2) and this makes it sound bigger (instead of running it on 18V)

[R.G.]

Wouldn't the increased voltage change the way the 308 clips before it hits the diodes?

The diodes start clipping at about 0.6V above/below bias . The 308 doesn't clip til its output voltage hits the power supplies, less a volt or two. So the diodes are clipping long before the 308 can start, even at 9V.

I do remember reading that the limited slew rate of the op amp used in a Rat is part of the sound - by doubling the voltage, does this mean that the op amp takes twice as long to swing through the full range?  Might this change the sound a little?

The LM308 was an innovation in its day for having a much higher slew rate than the other then-available opamps. It is possible to slow it down further, just as it's possible to slow any opamp down. It's possible that the UN limiting of the slew rate on the Rat is part of the sound, compared to other similar circuits using (for instance) the 741 or 1458. In either case, slew rate limiting is a large signal effect, and again, the diodes clip before you get to what's normally considered large signals. Techie words like "slew rate limiting", "transient intermodulation distortion", and various combinations of "phase" with other things do make great advertising copy, though.

Mustn't forget the psychological aspect of modding - if you want it to sound different, it will. 

Yep.

maybe he changed the input cap value (x2) and this makes it sound bigger (instead of running it on 18V)

Very definitely it will. Also the "changed some other caps" will.

Don't get me wrong - it's possible that there is some effect I don't know about that changes the sound. But my experience is that capacitor values have a huge effect on sound, and I don't see how changing the voltage does.

[StephenGiles]

Ah but can you play faster now??

[PRR]

> LM308 was an innovation in its day for having a much higher slew rate than the other then-available opamps.

You may be thinking of LM318 ?

The 308 "are precision operational amplifiers having specifications a factor of ten better than FET amplifiers over a .55°C to +125°C temperature range.
"The low current error of the LM108 series makes possible many designs that are not practical with conventional amplifiers. In fact, it operates from 10 M source resistances, introducing less error than devices like the 709 with 10 k sources. Integrators with drifts less than 500 uV/sec and analog time delays in excess of one hour can be made using capacitors no larger than 1 uF."

The claim was Low Input Current Errors. Also low supply current.

Slew is not quoted, though typical Large Signal Frequency Response (Power bandwidth) is. With stock (LM101) compensation the output exceeds 0.6V (diode-clip) up to 50KHz (though when allowed/forced to swing 14V it quits just past 1KHz). GBW with 30pFd comp is 300KHz.

OTOH, the 701 and its goofy compensation, at high gain, can slew many dozen V/uS and GBW in excess of most general-purpose opamps 20 years newer. In any speed race, a de-comped LM301 will whup a LM308.

The LM318 boasted "a factor of ten increase in speed over general purpose devices". GBW 10MHz, PBW at 13V to 1MHz. Costly and unforgiving.

I'm not sure why the LM308 is favored. Surely the low supply current was an attraction when pedals ran on batteries, but that day is fading. Perhaps the low-low-current input stage can not slam the amp as hard as "better" opamps do, leading to less violent clipping, although the Pulse Response traces show clean not-soft action.

[ashcat_lt]

I agree that too many variables were changed at once for the experiment to be meaningful in any pseudoscientific way.

I do wonder why the opamp clipping is being discounted.  Sure, the diodes are clipping at a lower voltage than the opamp, but they're doing that after the opamp, not before.  There's plenty of gain in the circuit to slam the opamp into it's rails.  Lifting the clipping diodes will show that there is already some distortion happening at high settings.  It does get clipped again by the diodes, but I have a hard time believing that the opamp clipping first has no impact on the sound.

Then there's the fact that if the opamp's output is bigger, the diodes will be clipping earlier, and harder.  That is (proportionally) closer to the zero crossings and for a longer portion of the cycle.  That will sound like more distortion, of course, but it will also increae the RMS output some, making it sound somewhat louder.

[anchovie]

Then there's the fact that if the opamp's output is bigger, the diodes will be clipping earlier, and harder.

Changing the supply voltage has nothing to do with closed-loop gain. The gradient of any section of the waveform within the clipping threshold of the diodes will be the same, so there is no "earlier" or "harder" clipping from them.

[R.G.]

I do wonder why the opamp clipping is being discounted.  Sure, the diodes are clipping at a lower voltage than the opamp, but they're doing that after the opamp, not before.  There's plenty of gain in the circuit to slam the opamp into it's rails.  Lifting the clipping diodes will show that there is already some distortion happening at high settings.  It does get clipped again by the diodes, but I have a hard time believing that the opamp clipping first has no impact on the sound.

It doesn't get clipped again by the diodes; it gets first by the diodes and later (in time) by the opamp.

Here's the rationale:
- the signal can be consider to be rising or falling from 0V, as it does crossovers of 0V
- as it rises, it does so with some slope - that is, it does not instantly jump from 0V to maximum; this is certainly true with a guitar going into it; the signal takes some time, some measurably milliseconds to rise per volt, so it gets to higher voltages later in time
- as it rises from 0V, there is no clipping until it rises to the diode clipping level
- since the diodes are not letting current flow, the voltage on the diodes tracks the output of the opamp closely
- the diodes start to conduct, clipping the signal at the output
- the opamp output signal continues to rise after the diodes have started limiting the output signal
- some time later, the opamp clips; by that time, the diodes are already into hard limiting, so what the opamp is doing is well covered up by the fact that the diodes started clipping the output signal long before the opamp did.

The fact that the opamp processes the signal before the diodes get it is overruled by the much lower voltage at which the diodes clip.