CAPI CA-0252 opamp

Now you made me want to spend more money 💰 😩

Is there a layman’s terms idea for what all the opamps sound like? I’ve got red dots in my vp28’s and love them. But... what would I gain from switching it up to something else? Hope I’m not beating a dead horse here.

I’m about to build two bt50 and two fc526

I’ve built several of the gar1731 and 2520 in the past. It’s very tedious work to me. Am also thinking about diyre red 25.

You’ll have a much easier time building the 0252. The likelihood of solder bridges is much less IMO.

Hahaha I bet there are some things.

Dag nabbit. That means I need to make some.

I have some thoughts on this based on the opamps I've built. For reference, I'm using these in CAPI VP28s, LC25s, BT50s, ML2s, and a pair of Analog Allstars EQP1S5s. I haven't tried every possible combination, but I have done quite a bit of swapping on the bench with full program material on loop to decide what I like best.

In order of least to most vintage sounding in my totally worthless and extremely subjective opinion:

Modern (cleanest, sparkling highs and tight lows)
gar2520
garOA10
gar1731
garAM10
ca0252
Vintage (more distorted, smoothed highs and wooly lows)

Now, I am at least partially in agreement with IAA Matt above with regards to swapping DOAs; a single amp often doesn't make a big difference and it's highly dependent on what you're plugging it into. That said, A) we're all here because the finer points in audio matter to us more than most, and B) the sounds of our favorite classic recordings came in part from running multiple tracks through multiple amplifiers. The small things do add up. There are so many other things that will make a much bigger difference in the recording, I can understand the viewpoint that a single opamp is insignificant. Given the low cost of trying out a few options though, I think it's worth it for the confidence that comes from knowing you've made the gear sound as "good" as it possibly can and gone the extra mile to craft a signature sound.

The CA-0252 in particular is special IMO. First of all it's another top quality CAPI product and the easiest to build of the ones I've tried. It has the most vintage character and overall difference in sound by far. It's straight up rock n roll is what it is. While it's great, it might be too much depending on your application. Trying it against 2520 and 1731 in the EQP1S5s, it was doing too much for me and I settled on the 1731 (still need to try OA10 in those, my favorite in my other eqs). I want to leave the option open to use those eqs on the mix. So consider the application and how many stages are in your overall chain. 0252 would be my first stop for trying to get maximum color out of the shortest chain.

I have resigned myself to the fact that I do not have golden ears so I have to rely on testing - so this is based off of an empirical view. The conventional rule of thumb for distortion to become audible is 1% total harmonic distortion plus noise. However, whether you can hear it or not is extremely dependent on the frequency of the fundamental and the harmonic of distortion. In tests people often cannot hear low frequency distortion even when it is at the same level as the signal - we are practically deaf to distortion below 40Hz, but most can hear higher frequency distortion below 1%.  Our sensitivity to distortion is also related to the complexity of the signal, and there is a masking effect from the fundamental which means 2nd and 3rd order harmonics aren't often as perceptible as 5th or 6th, for example (which contributes to the previous observation).

No well-functioning op amp in most cases produces higher-order distortion products (like >3rd) above what would normally be considered perceptible levels. All of the op amps on your list have >50dB open loop gain at 20kHz, which means at normal operating conditions the distortion they produce will be reduced by negative feedback by a significant amount, and all of the op amps on that list have a strong general distortion profile of each order of distortion being lower than the previous - so masking and whatnot are in play.

Slew rate differences in most discrete op amps aren't relevant at normal voltages within the audible range. Just by raw numbers at normal listening levels, and nominal voltage levels (meaning around +4 dBu) there are few measurable differences between op amps most of the time. The effects are (should be!) dominated by the circuit and components around it, with some exceptions as I mentioned before.

Which is as least part of the reason for the difference in sound when swapping op amps - the circuits are often optimized for other op amps. In the case of the AM10, for example, it doesn't really function properly until around +/-20V on supply (the DC open loop gain will jump from ~70 dB at +/-15V to ~120 dB at +/-20V and the distortion profile will change as well). So yeah, it'll sound very different from a 2520 at +/-15V... but they'd be much closer if you ran the same test with the 2520 at +/-15 and the AM10 at +/-24 or 28.

From my own testing I can tell you that your list doesn't line up with the distortion levels I've measured on those. The OA10 should have the highest THD, followed by the AM10, then the 1731, and then the 2520. But those numbers are dependent on level and load - the OA10 makes more distortion overall but the 2520 makes more odd-order products at higher levels.

Anyway, sorry, I'm rambling. Maybe they do all sound wildly different, and it's a world I simply can't participate in.

Take a scroll thru the build guide doc and see what you think. I worked hard to make sure it was the easiest build I have ever seen.

capi-gear.com/catalog/Build_CA-0252-DIY_Rev_A.php

Excellent build guide Jeff!! jsteiger !!  But that's a **** ton of stuff on a postage stamp.  Maybe I'll have Matt build a couple for me!!  Yeah, that sounds like a plan....