More on the INA217 etc....

OneRoomStudios

OneRoomStudios

New member
So I noticed in another thread that you can order free samples from TI, so I ordered samples of two different INA217's and an INA163. I'd like to to build the preamps that are diagramed in the datasheets on the TI website, but there are a few things I'm not so sure about. The diagrams for both models recomend using an OPA137 to drive the DC output to 0V. Which OPA137 should I use? Here are my choices:

http://focus-webapps.ti.com/general...nid=B1GONFHDFMSDVQC1JAWCM4Q?partNumber=OPA137

I think it would be pretty cool to difinitively see which chip sounds better (or if there's even a noticible difference). Plus two sets of high quality preamps can't hurt...especially when the chips are free. If anyone has any input or helpful hints, I'm all ears. Thanks,


-Peter
 
OneRoomStudios said:
So I noticed in another thread that you can order free samples from TI, so I ordered samples of two different INA217's and an INA163. I'd like to to build the preamps that are diagramed in the datasheets on the TI website, but there are a few things I'm not so sure about. The diagrams for both models recomend using an OPA137 to drive the DC output to 0V. Which OPA137 should I use? Here are my choices:
Click to expand...

If you are going to use it, probably you want OPA137PA.
 
try this as an alternative to the ina approach.
if the opa137 is a varaiety of the standard op27/37 .
which i suspect it is.
try the opa137 as a first stage configured as a differential input op amp.
using the LT1115 pdf mic pre schematic at linear technologies web site.
then take the output of this circuit to a line mixer to control volume going to the sound card.
voila ! fewer parts and complexities. no INA needed.
 
well it really depends hilarious. technical op amp specs ive found sometimes bear a different relation to how the final product sounds.
there are op amps around with 1nv noise, but a lot of folks - me included , some times prefer some other op amp with a worse spec because we like the sound better.
5532's - not the best spec op amp were used in a lot of hi end stuff for years.
all i'm suggesting is an alternative. and to my ears frankly i like the
max 437 as ive said before. ive never tried the opa137. but it IS a fet op amp. so i would be surprised if it didnt sound good. particularly from my checking it seems to be built for audio circuits.
peace.
 
also a neat idea might be to try a llundahl transformer or a jensen
in front of that opa137 and get a bit of iron in the equation.
could be an interesting alternative.
 
manning1 said:
well it really depends hilarious. technical op amp specs ive found sometimes bear a different relation to how the final product sounds.
there are op amps around with 1nv noise, but a lot of folks - me included , some times prefer some other op amp with a worse spec because we like the sound better.
5532's - not the best spec op amp were used in a lot of hi end stuff for years.
all i'm suggesting is an alternative. and to my ears frankly i like the
max 437 as ive said before. ive never tried the opa137. but it IS a fet op amp. so i would be surprised if it didnt sound good. particularly from my checking it seems to be built for audio circuits.
peace.
Click to expand...

I'm with you one the 5532, but the OPA137 is described as "Low Cost Operational Amplifier" on the TI website, versus 5532 which is "Low-Noise", INA217 "Low Noise Low Distortion", INA163 "Low Noise Low Distortion" and others like OPA134 "High Performance Audio" and OPA2604 "Low Distortion". Methinks TI is trying to tell us something.

I think the OP already has both from free samples . . . sure, if there's enough to go around, might as well see if 137 makes an interesting colored pre, I guess.
 
Just logged on to the TI site... ordered 4 INA217's & 4 OPA137's... they're being shipped overnight and not one red cent out of my pocket...

Sweet... thanks guys
 
hilarious. I KNOW it seems weird. but i HAVE tested a LOT of top op amps.
over many years, and so called very low noise transistors.
the op137 is probably one of the few ive never tested.
heres some conclusions i came to quite a while back.
discard and call me crazy if you wish.
1. i use a standard op amp test bed. so i can plug in different op amps.
what i do is actually record tracks with the new op amp rather than
going mad on the scope approach. this tells me something no scope can tell me. ie: the texture of the sound. its character.
2. ive found some of the lowest noise op amps sometimes sound , how can i put this tactfully "clinical". like a hospital room. as the name implies often "instrumentation amplifiers" are often designed to fullfill very accurate instrumentation needs rather than the texture i look for for audio.
for example the aerospace industry and others where very accurate measuring devices are needed.
3. i found in my sound experiments that current noise is more important than
voltage noise in the specs of an op amp. and seemed TO ME to definitely play a part in me getting a better sound. ie: the lower the current noise in the fa spec the better i liked the sound. if you notice BB have some quite
low current noise op amps.
4. one night - which is rare for me i got drunk and inadvertently loaded the wrong transistor into a mic pre i was building. just to see what would happen. compared to the super low noise high price transistor i had been using. frankly this tranny was a hunk of junk i would never normally use , but somehow it had got mixed in wth the good pile. this taught me a good kick in the pants lesson. on recording tracks i was blown away by the sound and its character. then stupidity took over. to this day i cant believe i was dumb enough in my stupor to rip the test rig down before i fell into bed.
my wife laughs about it to this day. RULE - dont forget to document ones great designs that work. ive never been able to recreate it sadly.
so i concluded after this experience there is more than specs to building a great mic pre.
all the best.
 
manning1 said:
hilarious. I KNOW it seems weird. but i HAVE tested a LOT of top op amps.
over many years, and so called very low noise transistors.
the op137 is probably one of the few ive never tested.
heres some conclusions i came to quite a while back.
discard and call me crazy if you wish.
1. i use a standard op amp test bed. so i can plug in different op amps.
what i do is actually record tracks with the new op amp rather than
going mad on the scope approach. this tells me something no scope can tell me. ie: the texture of the sound. its character.
2. ive found some of the lowest noise op amps sometimes sound , how can i put this tactfully "clinical". like a hospital room. as the name implies often "instrumentation amplifiers" are often designed to fullfill very accurate instrumentation needs rather than the texture i look for for audio.
for example the aerospace industry and others where very accurate measuring devices are needed.
3. i found in my sound experiments that current noise is more important than
voltage noise in the specs of an op amp. and seemed TO ME to definitely play a part in me getting a better sound. ie: the lower the current noise in the fa spec the better i liked the sound. if you notice BB have some quite
low current noise op amps.
4. one night - which is rare for me i got drunk and inadvertently loaded the wrong transistor into a mic pre i was building. just to see what would happen. compared to the super low noise high price transistor i had been using. frankly this tranny was a hunk of junk i would never normally use , but somehow it had got mixed in wth the good pile. this taught me a good kick in the pants lesson. on recording tracks i was blown away by the sound and its character. then stupidity took over. to this day i cant believe i was dumb enough in my stupor to rip the test rig down before i fell into bed.
my wife laughs about it to this day. RULE - dont forget to document ones great designs that work. ive never been able to recreate it sadly.
so i concluded after this experience there is more than specs to building a great mic pre.
all the best.
Click to expand...


That's cool. I'm never used 137 either, mostly I rip out cheap crap like 072s & replace with 134 series. And when I want flavor I use tewbs :D
 
I built one

I built up a preamp using a Lundhal input transformer, INA217 and OPA137. It's very transparent, black box straight wire gain. No input caps, no output caps. The OPA137 is used to null out the DC offset from the 217. That's all. I'm very happy with my preamp. Sure it's clinical, no color but clean and dynamic. I can add whatever color I want later in software. There may well be differences in color from chip to chip but I don't think the difference will be way huge. Go for clean, quite, dynamic, low distortion. Add your favorite colors and fix'ins later in software I say!
The less you do outboard the better. That means less caps and devices. Software is getting cooler and cooler all the time. Add your effects and colors to dry, clean tracks.
Oh and, I might add, the OPA317 has no effect on the audio.

Bob
 
hilarious. a TIP. more than a few people have told me they like the analog.com replacements for the standard 072's. i think - but i'm not sure AD712 comes to mind. might be worth getting a sample to try.
i hope i got model no right.
anyway just ask adi for the drop in replacement for the 072.
they have a toll free number for tek q's. on their web site.
also another TIP - try and get the huge analog devices book .
this is a superb op amp book showing loads of applications including audio.
definitely a must have book. they normally send it free via the 1 800 number if you call them and are nice. youll learn a ton from this book.
also linear technology have a nice set of books as well.
including some mic pre designs.
 
bobs mods. if you get time. try the llundahl into a max 437 or HA5137 or 47 or LT1028 op amp stage.
tell me what you think.
 
The INA217 is socketed but is an instrumentation amp. I couldn't easily plug in op amps for comparison. What is it about these devices that would be worth the time and trouble to A/B them?
I expect semicondutor devices will differ mostly around the dynamics and noise, unlike differing tubes that have distict "flavors".

Bob
 
manning1 said:
4. one night - which is rare for me i got drunk and inadvertently loaded the wrong transistor into a mic pre i was building. just to see what would happen. ...
...on recording tracks i was blown away by the sound and its character.all the best.
Click to expand...
Maybe you were just drunk. I too have had some amazing experiences in an altered state of consciousness. :)
 
doc. nah - i thought about that at the time.
the next day when i was sober. i listened back again and many more days after. still i liked the tracks better.

bmods. well it was just an idea. no you cant just drop in a max437
to replace the IA. actually ive found quite a lot of different "colors"
between op amps. so i have to politely disagree.
in fact ive even done crazy stuff like plug in a op amp not made for audio.
and several times found interesting flavors. just my observations.
ive been toodling around with op amps for many years and learnt the hard way that specs dont tell the whole story. but to be frank i just dont like the 5532's sound. just personal preference.
 
I wouldn't doubt your findings for a minute. Nothing tells the truth better than a working prototype. Specs are part of the story, the listening test is the rest of the story. Differing internal layouts and engineering design can account for changes in color, just like differences between like tubes. My expectation is that semiconductors should not exhibit as much as a color difference as tubes or transisters would.
Hell, theory tells us that recording at 44.1 kHz is all that should be needed but recording at 96 khz and converting to 44.1 kHz does produce better tracks. I suspect there are hidden phase and rolloff issues at 44.1 kHz that are not fully understood or disclosed.
My interest is colorless sound and I found that with the INA217 so I've not ventured further. I also use the OPA2134 as a substitute where I find those JRC4850s which are found like grains of sands on beach of music gear.
 
bm.i wouldnt touch a jrc if i was given one.
however i would suggest people using transistors look at zetex e line if i remember. i could never blow them up.
 
Bob's Mods said:
I built up a preamp using a Lundhal input transformer, INA217 and OPA137. It's very transparent, black box straight wire gain.
Click to expand...



You wouldn't happen to have a schematic for that would you? :) How close is it to the setup in the TI datasheets? I'm definitely with you on the whole colored pre thing. I'm sure there are pre's out there that sound beautiful, but I just want my sound going into my comp clean; I can fuck with the sound once it's in.

-Peter
 
You must log in or register to reply here.
Back
Top