Don't go changing opamps willy nilly

[JohnRoberts]

there's the other side of the coin , especially in the building construction trades , to get from contractors
" ah , you'll never notice , or everyone does it that way ,  It doesn't make a difference or [ insert your own ,
and indeed , you get it inserted !  ]

usually , cause they want to get out as quick as possible with as much profit, with the least amount of effort
there is an element of this in every industry

I have been involved in product design for several decades and there are some well established axioms or universal truths.  Hidden costs that the customers do not perceive are worse than leaving money on the table, you make the SKU uncompetitive compared to other similar designs at point of sale that did not add the extra (hidden) cost.

When I first started designing products at Peavey in the mid-80s I had the luxury of being such a low cost manufacturer I could afford to put some extra stuff inside products mainly to please myself. By the 90's competition and Chinese manufacturing erased that low cost competitive benefit, and I had use a sharper pencil for my design/features (anything that cost money is a feature).  This is not good or bad, it just is. 

Long ago I learned to watch what customers do and ignore what they say. If customers actually did what they say, Behringer never would have enjoyed record sales from making Chinese versions of popular domestic products. Even Mackie was hoist on their own (petard) "Made in USA" marketing campaign. 

When I was over the product marketing  department I recall the difficulty my transducer sales manager had marketing the field replaceable basket for peavey speakers. It was an extra "hidden" cost that many customers did not appreciate when making price comparisons. It is hard to market a feature to a customer with a sentence that begins "when your speaker breaks....". 

JR 

PS: Speaking of trades I am constantly reminded of corners cut by the contractors who worked on my current house. Hopefully they are dead and gone by now.

 

[Tubetec]

Interesting insights from industry as usual John ,
And damn those peavey Bw's with the replacable baskets was a fantastic idea ,after a blow out you could have your bins up again in a matter of minutes .

 

[okgb]

PS: Speaking of trades I am constantly reminded of corners cut by the contractors who worked on my current house. Hopefully they are dead and gone by now.

Especially in the trades , you literally  have to " live with it "  most of those guys seem so , take the money & run
that  I'm ok to pay more when I feel there's honesty , good service & product

and then there's the subjective things like instrument set ups  ........

 

[JohnRoberts]

Especially in the trades , you literally  have to " live with it "  most of those guys seem so , take the money & run
that  I'm ok to pay more when I feel there's honesty , good service & product

and then there's the subjective things like instrument set ups  ........

The young local electrician that I was cultivating to perhaps clean up some dodgy wiring around my house, didn't return my calls over a few week period... in fact he still hasn't returned my calls.

I did the wall oven tie in myself. In fact I was more interested in getting a strong back to help lift the oven (110# ship weight), but I got er dun myself, because I had to (and I could).

JR

 

[okgb]

I have a car mechanic like that ,  he often talks me out of having something done
I never question his bills!  I value and try to keep good relationships with reasonable quality people
and as the thread started , companies making good valued & quality products

 

[user 37518]

I'm ashamed to admit but, I sure do love me a Tl064 or a 4741 on some inputs....... Wish it...064... wasn't as noisy but it just works where I have them...... At least it leaves my small LME49740 stock for when I figure out what I'm doing....... :-[

The TL064 1MHz unity gain bandwith is no good,  the 4741 has a 3.5 MHz GBP but a lousy 1.6 V/usec SR. What are you using these chips for?

 

[JohnRoberts]

I used a TL064 in my first generation battery powered drum tuner because of the lower PS draw, and almost normal op amp behavior. Back then the still new very low current op amps I tried acted squirrely the way I tried to use them.

If you are not running from battery power dropping in TL07x op amps seems an easy and logical upgrade.

JR

 

[user 37518]

They are already in an older compressor and line splitter ..... TL Audio Fat1 and a Rane SM26........
I've tried "upgrading" them and ended up going back to what was in there more or less.... I think that, without really redesigning or modifying more than adding some of the  bypassing like I did , there's not too much to gain from better op amps in these designs . There are obvious differences I can hear but, the qualities I like to have available most of the time seem to weigh more on the original op amps..... More comfortable ..... Not transparent but, not bad at all for use in other things.....
I can goof around with different impedances and levels feeding different things and get things to sound great/different but, it's too much thinking sometimes and I'm dumb.... The 064 and 4741 have a way of taking different input signals and are more forgiving if that makes sense... Sorry for the phoolish description....

I agree, when things work they shouldnt be messed with, if one is designing something from scracth you can select the part you like, but we should give some credit to the original designers, however like JR pointed out, sometimes the selected part by the designers has more to do with cost rather than performance, in which case it may benefit from judicious replacement.

 

[JohnRoberts]

I agree, when things work they shouldnt be messed with, if one is designing something from scracth you can select the part you like, but we should give some credit to the original designers, however like JR pointed out, sometimes the selected part by the designers has more to do with cost rather than performance, in which case it may benefit from judicious replacement.

I don't remember saying that...  Back when I was buying truckloads of op amps I paid embarrassingly small prices for decent performance stuff.  I used decent audio speed op amps for audio paths, and cheaper than dirt GP op amps for non audio glue, like peak indicators or whatever.

My recent comment about using a TL064 was because of its lower current draw, not cost.

JR

 

[user 37518]

Yes.....

I was kinda hoping for someone to tell me that duh, change Rx and Ry to this value, try this value for c? and then try the LmE at ic?? and see what you think.....lol....Just kidding....

I don't know, I tried the 074, the 347 and the 49740 at the input and it just didn't work as well as the 064 with how I was using it..... I'm sure it has more to do with how I had things going on with my set up and goal than whether or not the 074 is superior overall..... I do think everything went to the 074 in later models without any changes made from what I understand.....it's even on the schematics....

It is interesting to hear the changes when swapping ics.... It's gotten more obvious, like learning to hear compression...... It took me about 10 years of life to understand what I was hearing.... but, it's not my main gig so I don't consider that abnormal maybe.... 

These op amps are quite impressive as a whole I must say. What a cool thing to be able to have the technology advance the way it has and even cooler to think that so much has gone into using them for people to enjoy music....Fascinating.... I Love It!!!! And I Love this place!! 

What about measurements? have you measured the equipment after performing those changes? or are you just relying on ear? Measuring thing out helps putting things into perspective,  THD, IMD, frequency response, even an FFT helps you visualize things and see how it changes.

 

[JohnRoberts]

Yeah I just listened.

I was more or less just throwing it out there how I personally liked the 064 and 741 on some inputs because, for whatever reason, it just worked..I could be more sloppy or careless in my tweaking because it's like a big mess of not accurate but smooth if that makes sense...(how's that for some poetic phoolery)... and  It just so happened that they were the original design . I'm not sure why I didn't like the different ones I tried, maybe I don't like accuracy in this context or maybe there were some things about the circuit that didn't jive.??? They would clip differently and react/clip differently at different times etc.......

I guess it would be cool to see what was going on to find out. I'm sure  that's how Jim Williams makes his decisions when modifying circuits......... with test gear.... I know he's after keeping true to the source....

I figured out decades ago that I could measure stuff that I couldn't hear, and could measure anything I heard, so relied upon bench measurements to refine designs.

That said there are different philosophies for design. From the clean pot school ("always start cooking each meal in a clean pot") the linear path is preferable, with effects (distortion) added to taste (like seasoning).  Alternately the "if it sounds good it is good" (for all the time), just do what sounds good to everything, until it stops sounding good to you.

As  a manufacturer I couldn't presume to know what effects (distortion) other people will always think sounds good, and haven't designed studio efx since the 80s.  These days I don't even design clean audio paths, but technology has advanced to the point where off the shelf canned designs can match my better work of decades ago..

JR

 

[JohnRoberts]

I'd be looking for clean and transparent in conversion for the most part. Mic pres would be important for some too. I hate trying to capture something I'm hearing and it come out too different. But, if I'm creating or experimenting, I don't mind playing around with whatever..... it is what it is.......

To derail this a bit, regarding the Rane Splitter, I made a mistake and left an output op amp out of it's socket and noticed that the other particular channel I was using actually got louder and less distorted from what I can tell.... I then removed a couple of other op amps and noticed this more....... I didn't realize that the power supply had so much influence on the sound and performance of the chips.... interesting....

I'm guessing this is because the power supply is able to provide more current? It seems the circuit is able to function within it's different functions?? Weird....

Is this safe to continue operating like this????? Just curious..... I'd like to give that power hungry LME another shot ....

Also, any tips or resources I can find on what test methods are good for seeing what I'm hearing??? I have a scope and just my software SPAN for now..... How do you see or read distortion outside of squared signals????

It seems unlikely that a RANE design would have that marginal a power supply.  But it is not impossible for a power supply to suffer a component failure (or solder joint) over time.  Imagine if the reservoir capacitor is not smoothly providing current between charging pulses. That could show a load dependence.

Like wise a wiring fault or solder bridge that loads an output could behave like your symptom.

If you have a scope inject a clean sine wave and follow it through the path.... If the distortion returns when you plug the op amps back in look at that.

Look at the PS with your scope. Is it clean?

Its always something..

JR

 

[user 37518]

How do you see or read distortion outside of squared signals????

A THD meter works like a charm for this, If you can afford it, you can get an Audio Precision System One for about 1K, if not, there are cheaper options like the HP 8903 audio analyzer.  Some people also like measuring IMD since IMD is probably more "hurtful" to the ear than THD.

 

[JohnRoberts]

Interesting....I'll check it out....

Also My  particular 86 Rane has a tiny transformer and just a bridge rectifier and caps....No regulation I think...... ....... I can't find older schematics...... I need to peek inside this metal shield..... but, there's nothing past the bridge so I don't know if it's laid out different than what I'm seeing.....

Thanks!

Most op amps will work OK even with unregulated rails as long as the ripple voltage is moderate.

Something you can easily check with a scope..

If you can hear the distortion you can probably see it with a scope on a sine wave, while small amounts of low order harmonic are not very visible or very audible. I suspect you are talking about visible distortion.

JR

 

[user 37518]

Most op amps will work OK even with unregulated rails as long as the ripple voltage is moderate.

Something you can easily check with a scope..

If you can hear the distortion you can probably see it with a scope on a sine wave, while small amounts of low order harmonic are not very visible or very audible. I suspect you are talking about visible distortion.

JR

I've noticed I cant see on a scope anything below 0.5% THD, even at 1% THD the sine wave still looks pretty clean

 

[benb]

any examples other than the Neve BA-440 ?
what are the audible perturbations ?
are they measurable ?

I don't know of other analog examples offhand, but in the digital world of just a short half-century ago, there was the original TTL 7400 (74xx)  logic series. Then there was the faster 74Sxx and 74LSxx series - these were faster, mainly because a Shottky diode was placed between the base and collector so that, just as John described, current would conduct through the Shottky, keeping the collector voltage high enough so that the transistor didn't go into saturation, and thus turned off faster.

 

[JohnRoberts]

I've noticed I cant see on a scope anything below 0.5% THD, even at 1% THD the sine wave still looks pretty clean

It really depends on the nature or type of distortion...  Low order harmonic is not very visible (or audible) while say something like crossover distortions can be visible and audible at much lower levels. Of course it also matters what frequency the fundamental and harmonics are. Since human hearing sensitivity peaks in the midrange, we can hear the harmonics on LF tones more easily than same amount of distortion on HF tones. Hearing distortion on complex waveforms is much harder since harmonic overtones often occur naturally, IM and crossover distortion not so much, while I recall being annoyed by the added distortion from the early "aural exciter" vocal processing.

Back in the 80s I spent some bench time listening when designing the sine wave generator for the TS-1 audio test set. I think I finally decided on about a couple tenths of a  percent (low order caused by the AGC loop).  It was a trade off between even lower distortion and a much longer settling time. In strict A/B listening tests I could hear that distortion vs. it not present, but it was clean enough to be used to reveal speaker rubs and common driver distortions from broken speakers, and audio path problems.

JR 

 

[JohnRoberts]

My scope is acting up now too..... It needed a recap and, after putting some new ones in, my triggering is all whacky for some reason..... I didn't pull the front panel off so maybe there's something there..... Not too impressed with the TEK 2225 pcb material........ Layout is sweet though...... It's my first......

Tektronix is a well respected brand, but stuff happens

.....But it worked enough to see there really wasn't any ripple in the Rane power supply that I could see as an issue???........ power supply has been recapped too......

What's really crazy and, my other RAne is doing the same thing, is that all the op amps are only getting 8vdc when the board is populated with all of them. When I pull some op amps out, I can get close to 15v....... I need to put all of the original chips back in just to make sure that it's not something wrong with the unit....But both acting the same way leads me to believe the power supplies in them are weak....... can't handle much more than what was designed...... I need to look at what the transformers look like...They are in little shielded boxes....

That's maybe why I was not liking the LME as much as the 741 or 064 in these????? I was maybe starving the LME??? Would it getting only 8 or 9v cause it to clip easier?????

don't know, look at the data sheet for LME

Update;;;;;  Putting all the original op amps back gives everything 15v at the rails......... so, putting in the power hungry chips was pulling the rails down quite a bit.  I should've known this I guess...... Now I need to go check a few more things......

oops...  I suspect RANE engineered a number of different SKUs using the same power supply, and this was done before the modern current hogs existed. 

What are the consequences of running op amps at lower voltage even if the may be within operating specs???? Performance issues?????

again the data sheets should reveal acceptable operating voltages. I actually ran op amps off +/-8V rails one time when I had a particularly power hungry SKU (with digital processor inside). I had to full wave rectify my standard wall wart for more current, at half the voltage, and split that single supply for bipolar rails (and a high current +5V rail).  This was used in a -10dBV bedroom recording product so decent headroom above -10dBV nominal 0VU.

Another thing to look at in the data sheets is how close to the rails do your modern uber op amps swing? if they only swing to within 2 or 3 volts of the rails, that is even less actual output signal swing.

Sorry to derail this a bit......=Dual=

This seems the perfect example for why we shouldn't change op amps "willy nilly"...  8)

JR

 

[scott2000]

Sorry to add to this thread but I had a question regarding the same RANE splitter......

When I'm turning the channel output level pots, I get pretty big subsonic (1hz-30hz) AC jump..has to be at least 30-40db....maybe more.....just from turning the pots either direction....it's only subsonic and it only jumps then comes back down....I'm actually seeing this behavior when using a different piece of equipment's gain make up pot too.....

For example.....z6........It's coming from the pin 7 from what I can tell....5 and 6 are  fine there......and all the respective output pins from the other channels do this..........
If I add a big gain after the Rane, I can see my speakers oscillate......

This happens with no inputs or outputs connected as well..... from what I can see on the scope at pin 7........

It looks similar to about the 6 minute mark in this video when he's adjusting some bias pot........

https://www.youtube.com/watch?v=KtXE7s5DlcI

But I'm not using a signal.......



Rane schematic is in this thread too

 

[JohnRoberts]

I don't like the pot across the NF path of an op amp alone... Wiper bounce, or a faulty pot will result in no feedback, a bad thing for op amps.

JR

PS: I vaguely recall a design issue with one of those RANE products that required an engineering change... Have you checked with RANE?