Why are LM308s so expensive and kind of hard to find?

[GreySuits]

From what I have seen they run $8-$12 a piece and are not found in most places (like Small Bear).

Is there a good alternative?

[miketbass]

Demand - 10 years ago I couldnt figure out why anyone would bother to build or own a rat without the LM308. The chip had gone obsolete but I was able to buy them handily for $1 a pop IIRC. Never had a rat put into my hands for repair or mods that left the bench without a socket and the venerable LM308 in place.

Nowadays supplies have began to dry up and the pedal mod/boutique build craze have driven prices up. Consider the history of "vintage" JRC4558 chips and you will see that what was once a cheap and plentiful part used in tons of products is a buzzword magic mojo tone ingredient that players will spend good $$ for. Anyone sitting on a cache would be insane to not take advantage of this and trickle them out on ebay at 1000% mark up.

Unfortunately in this circuit, I am of the opinion that nothing sounds quite "right" except for the LM308. This is a hotly debated topic and many have claimed that there is no discernible difference between chips, especially when the compensation cap is adjusted for the chip. There may be some truth to the latter but I have spent countless hours with this circuit swapping chips and always came back to the LM308. The way the distortion breaks up with this chip past one o'clock on the dial is something special.

As far as subs many single channel opamps can be tried. OPA134, NE5534, and LM301 all sound good. Current production Rats use OP07 I believe. I would recommend installing a socket and socketing the compensation cap as well to play around. For myself, the last rat I built got a LM308 metal can from Small Bear. I figured if I was going to have to pay magic unicorn prices for a chip I may as well go all out - it was $7 or so I believe.

[Mark Hammer]

It would depend on what you want to use them for.  And if you want to use them for a Rat, it would depend on what aspects of a Rat appeal to you.  As for cost, they ceased production some time ago.  Why?  Because the gain-bandwidth product was poor, and better op-amps were developed.

At max gain/drive, the Rat circuit asks for gains into the thousands; way more than the poor chip is capable of, especially with a 9V supply.  Because of the very limited gain-bandwidth product (i.e., how much gain can be provided at any given frequency, sometimes called open-loop gain), it attenuates the upper mids and highs (i.e., it simply can't provide the gain at those frequencies) in a way that works for the pedal.

This thread - https://www.seymourduncan.com/forum/showthread.php?163048-The-LM308-OpAmp-and-the-difference-it-makes-in-the-ProCo-Rat-2 - suggests that precious little difference results from use of a different op-amp.  If you're stuck, I'd suggest use of a similarly "lousy", but cheaper and more available, op-amp - a 741.  NOte that the LM308 is externally compensated by that cap between pins 1 and 8.  The 741 requires no such compensation.

[smallbearelec]

Is there a good alternative?

This LM301

http://smallbear-electronics.mybigcommerce.com/op-amps-2/?sort=featured&page=2

is inexpensive, sounds good and actually Is an LM301. I add that last comment because the scarcity of many vintage chips has resulted in a flood of re-marked garbage in the broker and eBay markets.

[Mark Hammer]

Here is the gain-bandwidth product from datasheets for the LM308 and LM301.  Both can be compensated in the same way, but if I'm reading it right, the 301 has a little more gain available for mids and highs than the 308. Close, though. Whether that is audible in the context of a Rat circuit is a whole other question I can't answer.

[digi2t]

You can use LM108's as well. They're the mil-spec version of the 308. Or, the Russian K140UD14. I've used both, and they work quite well. I've sourced them from Ebay for between $3 to $5 each (not Chinese vendors).

[miketbass]

I was feeling compulsive and picked up a couple of the metal can Russian 140UD14 deals on ebay just now. Im curious to hear them, Ive never come across this part before. Hows the sound/similarities to LM308?

You can use LM108's as well. They're the mil-spec version of the 308. Or, the Russian K140UD14. I've used both, and they work quite well. I've sourced them from Ebay for between $3 to $5 each (not Chinese vendors).

[Rob Strand]

This LM301

http://smallbear-electronics.mybigcommerce.com/op-amps-2/?sort=featured&page=2

is inexpensive, sounds good and actually Is an LM301. I add that last comment because the scarcity of many vintage chips has resulted in a flood of re-marked garbage in the broker and eBay markets.

You might even be able to use a larger compensation cap to reduce the difference in sound.  The Rat has LM308 + 30pF across pins 1 and 8.   A LM301 + 100pF across pins 1 and 8 might come close to original.    You would obviously do better by hand tweaking the cap but that means having an LM308.

[GGBB]

You can use LM108's as well. They're the mil-spec version of the 308. Or, the Russian K140UD14. I've used both, and they work quite well. I've sourced them from Ebay for between $3 to $5 each (not Chinese vendors).

Also LM208 if you can find them. 108, 208, 308 are all the same chip meeting tighter tolerances. I once saw 208s right beside 308s for less than half the price.

[digi2t]

I was feeling compulsive and picked up a couple of the metal can Russian 140UD14 deals on ebay just now. Im curious to hear them, Ive never come across this part before. Hows the sound/similarities to LM308?

You can use LM108's as well. They're the mil-spec version of the 308. Or, the Russian K140UD14. I've used both, and they work quite well. I've sourced them from Ebay for between $3 to $5 each (not Chinese vendors).

I've only tried the K140UD14A, which is (on paper) the equivalent to the LM108/208. The "B" suffix has similar spec's to the 308.

As for sound, as it is with these things, it's pretty much subjective. In my best hippie poetry; the Russian version is a scant more ragged, less focused, but with tinges of harmonic content in different frequency ranges than the 108/308. Of course, the surrounding cast of characters (guitar, cables, amp) will play a role too, but all things being equal, it really comes down to wish cork you prefer to sniff. Either way, you'll soon have the opportunity to crack that sucker yourself, and have your own sniff session.

[Steben]

What about a TL070 wit fitting cap?

Late Rats have OP07

[Mark Hammer]

From a perspective of having the same pinout and using external compensation, it will "work".  To the extent that you view the "sound" of a Rat to stem largely from the clipping diodes, or from the op-amp properties AND clipping diodes, you'll get what you want...or maybe not.  The TL070 is a much higher quality op-amp than a 308.

But, see here: https://www.diystompboxes.com/smfforum/index.php?topic=113437.0

[Steben]

From a perspective of having the same pinout and using external compensation, it will "work".  To the extent that you view the "sound" of a Rat to stem largely from the clipping diodes, or from the op-amp properties AND clipping diodes, you'll get what you want...or maybe not.  The TL070 is a much higher quality op-amp than a 308.

But, see here: https://www.diystompboxes.com/smfforum/index.php?topic=113437.0

or here
https://www.diystompboxes.com/smfforum/index.php?topic=55234.0

[GGBB]

What about a TL070 wit fitting cap?

Can you find TL070s? I've not been able.

[Fancy Lime]

I think it is fair to assume that if the old Rats didn't use the LM308, it would cost a fraction of what it does these days. Luckily there are indeed alternatives. I have done a little experimentation with the most popular alternatives, which resulted in my personal ranking (from best to worst):

LM301
LM308
NE5334
OP07

The LM301 and LM308 are so close that to my ear it makes no real difference unless you compare them directly in the same circuit. And even then I can only hear it if I take out the clipping diodes (best Rat mod ever in my opinion but that is beside the point). The LM301 is also slightly less noisy. The NE5334 needs a significantly larger compensation cap and is NOT PIN COMPATIBLE with the LM308. 60-100pF are a good starting point. It sounds very different and is much less noisy than the LM308 or LM301. Really a matter of taste. For high-gain sounds with the diodes engaged, I actually like the NE5334 best. Why anyone at Proco would think it might be a good idea to use an OP07 is beyond me. The compensation cap does nothing on this chip (see datasheet, although the new Rats seem to still have the 30pF cap, correct me if I'm wrong) and it sounds plain awful with the diodes disengaged. Again, a matter of personal taste but I really don't like it. With the diodes engaged it sounds kind of similar to the LM308 but definitely less so than the LM301. I did not even bother to try the TL070 because of how terrible I find the clipping of the TL072. Or any JFET input opamp for that matter.

Cheers and happy experimenting,
Andy


EDIT: I just remembered that I reported on my Rat experimentations here:
https://www.diystompboxes.com/smfforum/index.php?topic=121671.0;topicseen
And once we're at it: try the Big Muff clipping mod, it's delightful!

[Mark Hammer]

I wish I could find it, but there was a portion of an interview with the fellow who came up with the Rat in Art Thompson's "Stompbox" book, reprinted in an old issue of Guitar Player magazine.  I have about 20+ years' worth of the mag in the basement and am disinclined to thumb through them all at the moment.  The fellow notes that he mistakenly used a smaller ground-leg value than he intended to, turning the gain waaaayyyy up, and was pleasantly by this "woooo" undertone.  As I understand it, that results from the 308's limited slew rate and lower gain-bandwidth product.  Now, depending on how much you value the "woooo", a 308 may be essential...or not.

[amptramp]

The LM308 was designed for low input offset voltage and low bias current and was the first of the op amps to use super beta transistors for the input stage.  The noise performance was not great - it never gets better than 30 nV/(Hz)^0.5.  The frequency response was "old school" with limitations on audio gain at moderate gains.  We used a number of them where gain accuracy at DC and low (audio) frequencies was important and input impedances were higher than other contemporary op amps could handle.  The loss of gain at high audio frequencies works well with some effects.

On the good side, it operates between ±2V and ±20V rails, so that works well with the stuff we build.  The input common mode range goes from 1 to 1.5 volts under the rails and that is better than a TL071.

Not many people used them in their heyday and they were never cheap.  We could afford them in military products but we knew other better stuff was coming, so it had a "best before" date that accounts for limited production.

[willienillie]

My limited experience:  I've used NOS metal can Motorola LM308AH, and they sound quite different than the DIP chip.  Not necessarily better or worse, but very clearly different.  Later I picked up some new(?) National Semiconductor metal can LM308AH (are those no longer made?) and they sound exactly like the DIP version.  Small Bear has some NOS Fairchilds, I'm curious how those sound.

[bool]

OT: I once had motorola "MC" 1741 and 1747 (14pin dual "741") and they also sounded a bit different to other makes of 741 (that more or less sound similar).

But yeah; 308's are so expensive exactly because they are overpriced!

Get that? ha ha

[Elektrojänis]

How about trying one of the discrete op-amp designs floating around to Rat ciruit and then tweaking that discrete op-amp to get somethin interesting?