OTA Chip Compatibilty

[dfx]

Hi to everyone!

I've been looking lately at my box of things and saw one of my f/x build many years ago, around 1992, for a class project. Never housed it before and to my surprise, it has a LM3080 OTA chip. At first glance, i thought it was a LM308 and banking on it so i could make a RAT clone.

Closed scrutiny, it wasn't and I looked up on the specs only to find out its LM3080 and already obsolete from National Semiconductor.

Anyway, my question is, could this chip be a direct replacement for the CA3080 OTA chip as used in Dynacomp or Ross comp f/x?

I can see it has also the exact pin configuration with the CA3080.

Any feedback on this will be much appreciated.

[grapefruit]

Yes, it can be used in place of a CA3080.


Cheers,
Stew.

[ayayay!]

IMHO, it doesn't sound quite the same as the CA3080, but that doesn't necessarily mean it's worse.  Seems a little more lo-fi to me, but it will work. 

[Mark Hammer]

Calling an LM3080 "lo-fi" in relation to a CA3080 is a bit like calling a piece of food you found in a dumpster "cleaner" than a piece of food you found in a public ash-tray.  They are BOTH considered to be several notches below many other OTAs in terms of their signal-handling properties.  I'm not disparaging them, since many noble, serviceable, and dependable circuits over the years have used them.  It's just the differences between them are minimal compared to the difference between 3080 variants and other OTAs like the 3280, 13600/13700, 3094, 6110, 662, and more recent devices like the SSM and THAT chips.

Long story short: Don't sweat it.  Use it with a clear conscience.

[ayayay!]

Oh come on Mark, take it easy.  Did I not clearly say it doesn't mean it's worse?  I was just trying to impress that they don't sound identical as pin-for-pin replacements. 

Seriously folks, this is a forum after all.  Relax. 

[Mark Hammer]

Relaxation, now in progress.....

I don't doubt that extended critical listening might reveal some audible differences.  My point was simply that, in the grand scheme of things, it wasn't really anything to be concerned about.  LM3080s are quite acceptable, not THAT different, and he should have no misgivings about using one in place of a CA type.

Just clarifying how to interpret your comment of "Seems a little more lo-fi to me..."......with a little unnecessary bombast thrown in for good measure.

[ayayay!]

Fair Enough Mark, fair enough.  Have a great day. 

[StephenGiles]

.......and even less difference in sound with a fuzz box in the audio line!!

[dfx]

A million thanks to everyone..... and where is now that board i etched meant for a Dyna/Ross Comp? Hmmm....


BTW, another great find in my parts bin. I got two metal can type CA3080 OTAs.... oh my...

Got to fire that soldering pen now. 



Cheers!

[JDoyle]

IMHO, it doesn't sound quite the same as the CA3080, but that doesn't necessarily mean it's worse.  Seems a little more lo-fi to me, but it will work.

I've got to agree with Mark on this one - the difference between the two is minimal, in fact, it HAS TO BE.

The LM3080 is a 'second source' for the CA3080, meaning, somewhere along in the life of the 3080 chip, some goverment entity decided to use it in a design - this required that the part be available from more than one manufacturer in case the first were to go out of business. The only way that works is if the second sourced chip is EXACTLY the same as the original - the LM3080 is MOST CERTAINLY a pin for pin replacement - Uncle Sam demanded it be so.

Any differences between the two are most likely related to the quality control of National. RCA created the entire CA3xxx series of chips to use in their retail sales - televisions, radios, etc. Because of this they were MUCH more concerned with the overall quality of the chips and may (I have no actual idea, this is just logical inference) have done a better job with quality control and sorting than National does. RCA sold the line to Harris which kept it active and even extended it a bit; Harris sold it to Intersil which kept it alive for a while and then killed the entire CA30xx line.

Keep in mind that at National, the profit is almost entirely based on chip yield per wafer, so lowering the bar on noise when it comes to sorting automatically increases profit. RCA had to SPECIFICALLY worry about the final product the semiconductor was used in, not just the semiconductor itself, so they probably had more stringent standards.

The circuit interior to both chips is EXACTLY the same - and as far as ICs go it is EXTREMELY SIMPLE. This means that every part of the manufacturing process has greater overall effect on the outcome of the chip. Without a doubt the weakness of the chip stems from the need for complimentary transistors - latterally diffused PNP transistors, as used in the 3080, are, quite simply, horrible. Gains over 30 are near impossible, the breakdown voltages are lower, and noise is higher.

SO - the upshot is - both chips are the same, and no matter what the situation, it is prudent to audition several 3080s of ANY type before using them in an audio signal path.

In fact, if one were to take 10 CA3080s and 10 LM3080s and do a double blind test - I highly doubt anyone could tell them apart with any reliable consistancy.

Regards,

Jay Doyle

[Mark Hammer]

This is not meant to be a criticism of Jono/ayayay!, but it is a very common trap we fall into here (I've fallen into it myself many times I'm sure, and please feel free to let me know when I'm ever doing it again) when we unintentionally mistake what is observed in a few select instances with what is always observed in a particular context.  The classic case is the venerable Bg Muff Pi. Stories abound of this issue vs that one, yet few people have ever sat in a room with 50 copies of issue A and 50 copies of issue B, and verified in blind fashion, that something was always true/better/worse for A than for B.

Again, NOT a criticism.  I mean, how many of us would ever realistically have a chance to spend the day trying out 50 copies of a pedal using an LM3080 and another 50 copies using a CA3080 (or an RC44458 vs JRC4558, or 2N5088 vs MPSA18 or 2N3565, etc. etc), under optimal conditions (i.e., minimal time for auditory memory to contaminate things, blind rating/evaluation until all examples were assessed)...or even 10 copies of each?  It just tends not to happen in "real life".

So the conundrum we're left with is that we often have to make a determination, under adverse circumstances, of not only whether we are hearing difference X, but what to attribute it to.  Is it the device? the production run?  the manufacturer?  the era? (same as production run, I guess) the circuit it's in?  tolerances?  what?  WHAT?  WHAT THE HECK IS IT THAT'S MAKING ME HEAR THAT?

It's an epistemological challenge, to be sure.  I don't think we are collectively lazy about it, but we do need to be more mindful about when we might be perched at the edge of making incorrect attributions, or perhaps make more effort to qualify the attributions we make.  Personally, I think ayayay DID this.....at least enough for MY needs.  Still, it remains something we all want to be on guard for.  You know how rumours get started on this internet thingie.....   One statement on TGP is like an untended campfire in the Australian countryside.

[ayayay!]

Well, about 1 1/2 years ago, I did try out a butt load of LM3080s and a small handful of CA3080s in a... I think a Rocktron big crush or something like that?  It was a project where I got it for 5 bucks busted or something like that... Anyway I had that compressor pedal...

My experience with socketing those in and out was that I could, by ear, hear the difference.  If you think I'm smoking crack, so be it.  I know they're supposed to be the same.  A thought did however occur to me that it was really not important because maybe the batch I had of either one was not quite the same.  So in other words, no it was not purely scientific.  And also, it was not that one was "worse" than the other.  Just that in the Rocktron, things seemed less reactive to pick intensity with the LMs, hence the "lo-fi, but not necessarily worse" comments I made.  I know that on top of all that, that effect should have been the least noticeable of all, since they're supposed to be identical.  Does it really matter?  Only to me I guess.     G'night.