Hi,
You got there in the end.
"Accurate only goes so far, pleasant is usually better" I have been saying that for years. What good is "technical perfection" whatever that may or may not be, when you find you actually don't like listening to it. It's no good telling everyone that this or that recording doesn't sound good "because the system is so accurate it shows all the flaws in the recording". Thats rubbish IMO. A good and sympathetically designed system lets the music shine through in 95% of recordings. Which to me is what it's all about 🙂
Edit, I got round to showing how to remove SMD the DIY way after this and a few other threads
http://www.diyaudio.com/forums/showthread.php?s=&threadid=127924&highlight=
You got there in the end.
"Accurate only goes so far, pleasant is usually better" I have been saying that for years. What good is "technical perfection" whatever that may or may not be, when you find you actually don't like listening to it. It's no good telling everyone that this or that recording doesn't sound good "because the system is so accurate it shows all the flaws in the recording". Thats rubbish IMO. A good and sympathetically designed system lets the music shine through in 95% of recordings. Which to me is what it's all about 🙂
Edit, I got round to showing how to remove SMD the DIY way after this and a few other threads
http://www.diyaudio.com/forums/showthread.php?s=&threadid=127924&highlight=
exactly. 'true & flat' sound isn't all it's cracked up to be.
thanks for the link to the instructions again - nice post.
cheers,
adam
thanks for the link to the instructions again - nice post.
cheers,
adam
next mods
Hey Adam
now after u have your Motu done, i liked to know what you would finally suggest - 5532, 2227, 4032? i've found the National Op's like LM 4562 or the LME 49740 or TI's OPA2228 to be possible upgrades -
BUT i'm unsure about the (for me) strange +10/-6V supply voltage on the NJM2115 ( asymetrical, higher then spec. ?) - should i use "single supply"- types? Has anyone checked the symetry of the signals (sinewave) after the mod?
AND: does BLA use the 5532 on both, input and outputs?
thx for responses...
Mitsch
Hey Adam
now after u have your Motu done, i liked to know what you would finally suggest - 5532, 2227, 4032? i've found the National Op's like LM 4562 or the LME 49740 or TI's OPA2228 to be possible upgrades -
BUT i'm unsure about the (for me) strange +10/-6V supply voltage on the NJM2115 ( asymetrical, higher then spec. ?) - should i use "single supply"- types? Has anyone checked the symetry of the signals (sinewave) after the mod?
AND: does BLA use the 5532 on both, input and outputs?
thx for responses...
Mitsch
Hey Mitsch,
i haven't tried the 5532's myself. i did take a look at a bla mod, but i cannot remember exactly what they used. i think i wrote it down somewhere. they may likely have also changed what they used.
however, between the ths4032's and the opa2227's, my preference is for the latter. that being said, they both sound far better than the stock opamps and it's just personal opinion.
of the 3 you suggested, i would personally go with the opa 2228 from what i've read online from others' feedback. i haven't actually 'heard' any of the three, though i would assume the 2228 sounds close to the 2227.
really when it comes down to it, i would recommend try modding a few inputs and outputs with samples and doing some listening tests. when you come upon a device or combination you like, buy a whack of them and do your whole unit. OR you may like having a few different colours of sound to choose from on your unit and leave it full of different samples i suppose. [which also works great if you run out of money. ; ) ]
re:input/output. though the stock motu uses different opamps for the input and output stages, they are similar in the voltage they get in the motu and similar in specs. all the opamps i tried work as replacements for both. for example, i used the ths 4032's on the main outs, one set of inputs, and one of the mic pre's and they're all working great.
i don't have a scope to do any checking after the mod, just cross my fingers and use my ears.
i'd be curious what your thoughts are after your mod.
cheers,
Adam
i haven't tried the 5532's myself. i did take a look at a bla mod, but i cannot remember exactly what they used. i think i wrote it down somewhere. they may likely have also changed what they used.
however, between the ths4032's and the opa2227's, my preference is for the latter. that being said, they both sound far better than the stock opamps and it's just personal opinion.
of the 3 you suggested, i would personally go with the opa 2228 from what i've read online from others' feedback. i haven't actually 'heard' any of the three, though i would assume the 2228 sounds close to the 2227.
really when it comes down to it, i would recommend try modding a few inputs and outputs with samples and doing some listening tests. when you come upon a device or combination you like, buy a whack of them and do your whole unit. OR you may like having a few different colours of sound to choose from on your unit and leave it full of different samples i suppose. [which also works great if you run out of money. ; ) ]
re:input/output. though the stock motu uses different opamps for the input and output stages, they are similar in the voltage they get in the motu and similar in specs. all the opamps i tried work as replacements for both. for example, i used the ths 4032's on the main outs, one set of inputs, and one of the mic pre's and they're all working great.
i don't have a scope to do any checking after the mod, just cross my fingers and use my ears.
i'd be curious what your thoughts are after your mod.
cheers,
Adam
i saw an offer for the
OPA2134 at 3$ ea.,
the OPA 2132 for16$ or 1.20 ??
(not easy to find Audio- performance Opamp at reasonable prices)
between OPA2227 and the NJM 2115 the spec- differencies ar not that remarkable, but i really don't know about it's soundperformance. thats why i was on the OPA2228, but it's hard to find. (15$ea?) But i've found:
http://www.fixup.net/talk/topic.asp?TOPIC_ID=2016
i'm thinking about AD8066 on In's and OPA2134 on Out's.
after reading:
http://tangentsoft.net/audio/opamps.html
any idea where to find good priced 30 Pcs AD8066 ?
regards, Mitsch
OPA2134 at 3$ ea.,
the OPA 2132 for16$ or 1.20 ??
(not easy to find Audio- performance Opamp at reasonable prices)
between OPA2227 and the NJM 2115 the spec- differencies ar not that remarkable, but i really don't know about it's soundperformance. thats why i was on the OPA2228, but it's hard to find. (15$ea?) But i've found:
http://www.fixup.net/talk/topic.asp?TOPIC_ID=2016
i'm thinking about AD8066 on In's and OPA2134 on Out's.
after reading:
http://tangentsoft.net/audio/opamps.html
any idea where to find good priced 30 Pcs AD8066 ?
regards, Mitsch
hey man,
sounds like good stuff to me, but let your ears be the final judge.
those two (the ad8066 and opa2134) were on my initial list as well, but not available as samples i think. so that's why i ended up with the ones i have now.
for reasonably priced stuff i'd just go to digikey.com most likely. there's a few other sites out there too that have decent prices.
that being said though, i'd highly recommend trying to get samples first and seeing how you like them. texas instruments will ship free samples to you, for free - really. other companies will ship you samples if you pay the shipping.
as an aside, i did some vocal tracking yesterday. as it turned out, one of my mics sounded better alone with the opa2227 and the other sounded better with the ths4032. the differences were subtle, but noticeable with close listening. when i put it in the mix, i almost that changed things too, and i ended up putting both mics through the ths4032s. both opamps sounded much better than the njm2115 stock opamps.
all that to say, trying a few options may be beneficial before you commit. also, it won't benefit you to do the outputs if you don't use them much. but if you're using them as sends for outboard effects and the such, by all means they should be changed.
one final note. you may look around at opamps for video use - they tend to be fast and they tend to be cheaper. that's what a lot of diy's have done in the past. they're just high speed voltage feedback amplifiers with low voltage and low distortion.
cheers,
AT
sounds like good stuff to me, but let your ears be the final judge.
those two (the ad8066 and opa2134) were on my initial list as well, but not available as samples i think. so that's why i ended up with the ones i have now.
for reasonably priced stuff i'd just go to digikey.com most likely. there's a few other sites out there too that have decent prices.
that being said though, i'd highly recommend trying to get samples first and seeing how you like them. texas instruments will ship free samples to you, for free - really. other companies will ship you samples if you pay the shipping.
as an aside, i did some vocal tracking yesterday. as it turned out, one of my mics sounded better alone with the opa2227 and the other sounded better with the ths4032. the differences were subtle, but noticeable with close listening. when i put it in the mix, i almost that changed things too, and i ended up putting both mics through the ths4032s. both opamps sounded much better than the njm2115 stock opamps.
all that to say, trying a few options may be beneficial before you commit. also, it won't benefit you to do the outputs if you don't use them much. but if you're using them as sends for outboard effects and the such, by all means they should be changed.
one final note. you may look around at opamps for video use - they tend to be fast and they tend to be cheaper. that's what a lot of diy's have done in the past. they're just high speed voltage feedback amplifiers with low voltage and low distortion.
cheers,
AT
smitsch : While OPA2228 is would be a good choice for that application if you were designing from the ground up, it's a somewhat finicky opamp that may not play nicely if you just drop it in to an existing circuit.
Unfortunately OPA2228 and 2227 aren't available as samples, although their singles are. Depending on you application, it might be viable to use the singles, either on
http://cimarrontechnology.com/single-to-dualop-ampadapter-so8toso8versionpn031101a.aspx
or on
http://cimarrontechnology.com/single-to-dualop-ampadapterpn020302.aspx
depending on the footprint of what you want to replace.
If you go that route, you can also get OPA827 samples at the moment - they're new, and a bit unproven, but should be very good.
You might also want to consider OPA2107.
All that said, OPA2132 is a very good opamp, and readily available. In my estimation, 95% of the sound is the same between OPA2132 and OPA2227, and OPA2227 is about 95% of OPA2107.
I haven't personally tried the "gold standard" OPA627 yet, but there are those who say that the OPA2132 provides 90% of the performance at 1/10th (or less) of the cost.
For the inputs, I'd try LM4562 - very low distortion, and what many describe as a "sweet" sound. I'd put OPA2107's on the outputs.
Unfortunately OPA2228 and 2227 aren't available as samples, although their singles are. Depending on you application, it might be viable to use the singles, either on
http://cimarrontechnology.com/single-to-dualop-ampadapter-so8toso8versionpn031101a.aspx
or on
http://cimarrontechnology.com/single-to-dualop-ampadapterpn020302.aspx
depending on the footprint of what you want to replace.
If you go that route, you can also get OPA827 samples at the moment - they're new, and a bit unproven, but should be very good.
You might also want to consider OPA2107.
All that said, OPA2132 is a very good opamp, and readily available. In my estimation, 95% of the sound is the same between OPA2132 and OPA2227, and OPA2227 is about 95% of OPA2107.
I haven't personally tried the "gold standard" OPA627 yet, but there are those who say that the OPA2132 provides 90% of the performance at 1/10th (or less) of the cost.
For the inputs, I'd try LM4562 - very low distortion, and what many describe as a "sweet" sound. I'd put OPA2107's on the outputs.
oohhh kwak mod...
hey guys,
just posting cause i "finished" another mod to my motu.
i sort of 'hacked' elso's kwak-clock to to a clock mod on my motu.
the motu uses a tlc2933 to generate what might otherwise be generated by a crystal, so i couldn't just drop the kwakclock in as is. what i did was take the power conditioning aspect of the kwak clock and doubled the positive side of it. this gave me two independent very clean power supplies. i had to do the mod mentioned here to drop the voltage from 5v to 3.3v:
http://www.diyaudio.com/forums/showthread/t-17194.html
http://www.diyaudio.com/forums/showthread/t-17194.html
but after that, i 'just' isolated the 2933 from the internal power source and switched it for myne. it really, really helped. much more focused vocals, less high end glare, more thump and low lows. clean.
the tlc2933 is supposed to have a jitter of under 100ps according to the data sheet, but dirty power makes it a lot worse. so i reckon i can live with it as a lot better than it was, but not perfect.
i may, or may not, try the other half of the kwak clock later (that is the ad8561 IC) that i THINK is intended to clean up clock jitter a little more. however, i'm not sure how it will work with the signal coming out of the 2933 into the kwak clock and back out again... not that electronically savvy. What's more, i don't own a scope, so i can't really measure it very well either.
but a thanks goes out to elso for his design.
i don't have a website, but maybe i can figure a place or way to post pictures for others' sake.
cheers.
adam
hey guys,
just posting cause i "finished" another mod to my motu.
i sort of 'hacked' elso's kwak-clock to to a clock mod on my motu.
the motu uses a tlc2933 to generate what might otherwise be generated by a crystal, so i couldn't just drop the kwakclock in as is. what i did was take the power conditioning aspect of the kwak clock and doubled the positive side of it. this gave me two independent very clean power supplies. i had to do the mod mentioned here to drop the voltage from 5v to 3.3v:
http://www.diyaudio.com/forums/showthread/t-17194.html
http://www.diyaudio.com/forums/showthread/t-17194.html
but after that, i 'just' isolated the 2933 from the internal power source and switched it for myne. it really, really helped. much more focused vocals, less high end glare, more thump and low lows. clean.
the tlc2933 is supposed to have a jitter of under 100ps according to the data sheet, but dirty power makes it a lot worse. so i reckon i can live with it as a lot better than it was, but not perfect.
i may, or may not, try the other half of the kwak clock later (that is the ad8561 IC) that i THINK is intended to clean up clock jitter a little more. however, i'm not sure how it will work with the signal coming out of the 2933 into the kwak clock and back out again... not that electronically savvy. What's more, i don't own a scope, so i can't really measure it very well either.
but a thanks goes out to elso for his design.
i don't have a website, but maybe i can figure a place or way to post pictures for others' sake.
cheers.
adam
jive talkin...
well, after hours and hours of work, i was unable to get the ad8561 part of elso's clock to work. it would work sometimes, sometimes not - usually couldn't get the correct voltage out. i need ~ 1.5v (that's the signal level out of the tlc9233), but the voltage coming out of the ad8561 usually was around .6-.8v and the voltage boost by putting a 5k resistor between pin 1 and 8 did nothing. EXCEPT sometimes it would work, seemingly randomly. probably something simple that someone with more knowledge could figure out easily.
anyways, after all that, i spent a bit A/B ing my modded clock (with the more consistent voltage input) with the un-modded clock from another motu unit. alas, i think my long-hours and time investment shaded my initial listening tests. it's not really that much better at all. mildly smoother, mildly more focused, mildly more bottom end.
anyways, my recommendation might be to try a different clock mod if i were to do it again (or buy a decent clock). probably the thing to do is utilize some sort of jitter reduction chip OR make an accurate crystal clock with some way to switch between the two frequencies that the 2933 generates (45.1584 mhz and 49.152 mhz).
the 2933 switches between the two by being sent either ~1v or ~1.3v. it would be clever to integrate some sort of switch from this so that you could avoid manually switching the frequency. also, i think the smpte signal comes in to the 2933 as well, so it can't effectively be completely removed.
anyways, if anybody has any clever ideas as to why the ad8561 wasn't working right, that'd be appreciated. in fact, i don't even know if it positively effects jitter or not at all - maybe someone can tell me that.
for now, my project has ended. back to recording.
cheers,
adam
well, after hours and hours of work, i was unable to get the ad8561 part of elso's clock to work. it would work sometimes, sometimes not - usually couldn't get the correct voltage out. i need ~ 1.5v (that's the signal level out of the tlc9233), but the voltage coming out of the ad8561 usually was around .6-.8v and the voltage boost by putting a 5k resistor between pin 1 and 8 did nothing. EXCEPT sometimes it would work, seemingly randomly. probably something simple that someone with more knowledge could figure out easily.
anyways, after all that, i spent a bit A/B ing my modded clock (with the more consistent voltage input) with the un-modded clock from another motu unit. alas, i think my long-hours and time investment shaded my initial listening tests. it's not really that much better at all. mildly smoother, mildly more focused, mildly more bottom end.
anyways, my recommendation might be to try a different clock mod if i were to do it again (or buy a decent clock). probably the thing to do is utilize some sort of jitter reduction chip OR make an accurate crystal clock with some way to switch between the two frequencies that the 2933 generates (45.1584 mhz and 49.152 mhz).
the 2933 switches between the two by being sent either ~1v or ~1.3v. it would be clever to integrate some sort of switch from this so that you could avoid manually switching the frequency. also, i think the smpte signal comes in to the 2933 as well, so it can't effectively be completely removed.
anyways, if anybody has any clever ideas as to why the ad8561 wasn't working right, that'd be appreciated. in fact, i don't even know if it positively effects jitter or not at all - maybe someone can tell me that.
for now, my project has ended. back to recording.
cheers,
adam
maybe not as crazy as i thought. realized that the power in my building is really, really wonky. varies as much as 13v from standard. the motu is really sensitive to voltage - in the winter now, with extra power drain from heaters, sometimes it just wouldn't work at all!
also, rented a 'big ben' and slaved the motu to that - sadly didn't make it much better.
i've heard good imaging on my monitors before, but there's something wonky with my motu (maybe mine only) where it just absolutely refuses to come into clear focus with imaging. i reckon something may have gotten fried internally after my initial tests and that it actually did sound better, but does not any more. i mean, i think even a monkey should be able to hear the difference between a big ben clocked sound and an internal clock that has a lot of signal noise and a lot of jitter ( i hooked it up to a scope, and man is the internal clock wonky!).
there is another chip in the motu that distributes the clock - the acex 1k. unfortunately, it has these functions called 'clocklock' and 'clockboost' which correct an incoming clock signal that is faulty. the unfortunate part is that it's really pretty jittery - 250ps.
anyways, perhaps the acex 1k chip is the one that is throwing everything off and making the results unpredictable. i mean, the clock signal that i measured from the chip was so wonky that i figure the clocklock feature must be kicking in.
anyways. i'm trying to figure a way around monitoring with my motu now as it won't come into focus for me.
oh yeah, i did get the power thing sorted out. bought an AVR from a local computer store for $30 that does the job - works like a champ and keeps the volts stable within a volt or two of 120v now.
also, rented a 'big ben' and slaved the motu to that - sadly didn't make it much better.
i've heard good imaging on my monitors before, but there's something wonky with my motu (maybe mine only) where it just absolutely refuses to come into clear focus with imaging. i reckon something may have gotten fried internally after my initial tests and that it actually did sound better, but does not any more. i mean, i think even a monkey should be able to hear the difference between a big ben clocked sound and an internal clock that has a lot of signal noise and a lot of jitter ( i hooked it up to a scope, and man is the internal clock wonky!).
there is another chip in the motu that distributes the clock - the acex 1k. unfortunately, it has these functions called 'clocklock' and 'clockboost' which correct an incoming clock signal that is faulty. the unfortunate part is that it's really pretty jittery - 250ps.
anyways, perhaps the acex 1k chip is the one that is throwing everything off and making the results unpredictable. i mean, the clock signal that i measured from the chip was so wonky that i figure the clocklock feature must be kicking in.
anyways. i'm trying to figure a way around monitoring with my motu now as it won't come into focus for me.
oh yeah, i did get the power thing sorted out. bought an AVR from a local computer store for $30 that does the job - works like a champ and keeps the volts stable within a volt or two of 120v now.
thanks for your reports, adam ....possibly someone has to buy the BLA external clock and we rebuild it..?....
anyway. From what i see, it could possibly help to revise the PSU, for Better DC Performance, lower noise?
!! has anyone the complete scheme for the MOTU 2408 Mk3?? !!
other very weak Point: The Volume-Pot-Meters!! does anyone know HQ types, easy to replace (same size, design)?
this would be great....
stay clocked...
anyway. From what i see, it could possibly help to revise the PSU, for Better DC Performance, lower noise?
!! has anyone the complete scheme for the MOTU 2408 Mk3?? !!
other very weak Point: The Volume-Pot-Meters!! does anyone know HQ types, easy to replace (same size, design)?
this would be great....
stay clocked...
yeah, it would be good to replace the psu. it's just such a tiny box, that it's difficult. I think getting a good voltage regulator and power conditioner helps a lot.
i might guess that the BLA clock isn't all that different from the other DIY clocks out there (LC, Kwak, etc) given that it comes in a small box. I mean, regardless, you could do the same thing. perhaps the thing to do - which is too much for me - is take the kwak clock to the next level by adding a clock divider/distribution circuit and putting a BNC connector on it. i suppose you'd also have to mod the clock a bit to be able to have a crystal for both 44.1 (& 88.2) and 48 (& 96). really, all you have to do to keep up with every professional studio i've ever recorded at, is get a good 48k clock going. (every studio i've ever recorded at still records at 48khz/24bit - no exception. it almost doesn't make sense why they make the 192khz gear at all - i've never, ever seen that rate used. mostly, they just make sure they have great converters - the RADAR system by iz is common.) anyways...
back to making music.
cheers,
AT
i might guess that the BLA clock isn't all that different from the other DIY clocks out there (LC, Kwak, etc) given that it comes in a small box. I mean, regardless, you could do the same thing. perhaps the thing to do - which is too much for me - is take the kwak clock to the next level by adding a clock divider/distribution circuit and putting a BNC connector on it. i suppose you'd also have to mod the clock a bit to be able to have a crystal for both 44.1 (& 88.2) and 48 (& 96). really, all you have to do to keep up with every professional studio i've ever recorded at, is get a good 48k clock going. (every studio i've ever recorded at still records at 48khz/24bit - no exception. it almost doesn't make sense why they make the 192khz gear at all - i've never, ever seen that rate used. mostly, they just make sure they have great converters - the RADAR system by iz is common.) anyways...
back to making music.
cheers,
AT
First step is done, my motu 2408MK3 has the OPA2134 on Inputs (*) and Outputs.
ive replaced all op's on output-pcb and the 4x 3Pcs. belong to the input section.
(*) i'm not sure about these four 4680 Op amps located in between the AK Dac's (4 Pcs.) > Adam, what do you say? belong to what? i beleve -10dbu/+4dbv "switch" ?
my first impression is that the outputs are quite better, specially mid to high frequencies, the bottom is ok, but not remarkably better then the original.
i'm not sure about OPA2134 on inputs. sounds a bit colored and i miss some low-end punch compared to 2115....would it help to x-change the (*) Opamp with OPA2134, or should i go for my next champ, the AD8620?
---to be continued....
ive replaced all op's on output-pcb and the 4x 3Pcs. belong to the input section.
(*) i'm not sure about these four 4680 Op amps located in between the AK Dac's (4 Pcs.) > Adam, what do you say? belong to what? i beleve -10dbu/+4dbv "switch" ?
my first impression is that the outputs are quite better, specially mid to high frequencies, the bottom is ok, but not remarkably better then the original.
i'm not sure about OPA2134 on inputs. sounds a bit colored and i miss some low-end punch compared to 2115....would it help to x-change the (*) Opamp with OPA2134, or should i go for my next champ, the AD8620?
says the seeker...done yet? tell about?
---to be continued....
my second inpression (after some sleep, and some intensive checks)
- the (*) opamps belong to the OUTS and have to be replaced also.
- the inputs sound much better then i've described in my last post (was a faulty Tube- Phono- Preamp in chain) -all in all i'm happy with the OPA2134, they do a good job, sound is improved, natural, punchy and has better resolution.
- the Op's where dirt cheep @ 1$ each... so i decided to buy some AD8620, just wondering if those can do a remarkable difference...
- the (*) opamps belong to the OUTS and have to be replaced also.
- the inputs sound much better then i've described in my last post (was a faulty Tube- Phono- Preamp in chain) -all in all i'm happy with the OPA2134, they do a good job, sound is improved, natural, punchy and has better resolution.
- the Op's where dirt cheep @ 1$ each... so i decided to buy some AD8620, just wondering if those can do a remarkable difference...
he he. always good to sleep before testing.
not sure which opamps you were talking about. assume you mean 4580s and not 4680s.
are you working on a mk3 or mk2 828? mine is a mk 2. the board layout is probably the same though, i reckon.
you probably got the two close to main out 1 and 2 jacks. (those are the main outputs). then there's another by the smpte in out - think that's just for smpte signal - so not important.
there's another 4 by the input opamps - i think those are for stereo coupling or something like that, but don't know. i researched it and decided not to change them, but don't remember why and what exactly they are. the bla mod - at least earlier editions, can't speak for later - didn't change those.
then there's a couple on the board in other places. there's the two by the headphone jack - change those if you use the headphone jack at all. there's also two lm2162's or something like that for the mic pres. change those too and see what you think. i like the ths4032 on that one. not much character, but really nice.
overall, i really like the ths 4032. order some free samples from TI. they'll even pay the shipping. you can order five of them. you can also order a couple different versions - so you could order more if you'd like.
however, those ones are about $5usd each to buy if you get a good deal, so not quite the same.
another one to try - some people would frown - is the LM 5532. a lot of high end recording gear has been using that opamp for years. for example, my tascam DA78hr high def 24bit digital tape recorder uses them. i find them to be nice as inputs because they are just open and clear - really open, really clear. the same thing would make them mildly unpleasant as outputs though. they're also really chep - about a buck each.
cheers and happy recording.
AT
not sure which opamps you were talking about. assume you mean 4580s and not 4680s.
are you working on a mk3 or mk2 828? mine is a mk 2. the board layout is probably the same though, i reckon.
you probably got the two close to main out 1 and 2 jacks. (those are the main outputs). then there's another by the smpte in out - think that's just for smpte signal - so not important.
there's another 4 by the input opamps - i think those are for stereo coupling or something like that, but don't know. i researched it and decided not to change them, but don't remember why and what exactly they are. the bla mod - at least earlier editions, can't speak for later - didn't change those.
then there's a couple on the board in other places. there's the two by the headphone jack - change those if you use the headphone jack at all. there's also two lm2162's or something like that for the mic pres. change those too and see what you think. i like the ths4032 on that one. not much character, but really nice.
overall, i really like the ths 4032. order some free samples from TI. they'll even pay the shipping. you can order five of them. you can also order a couple different versions - so you could order more if you'd like.
however, those ones are about $5usd each to buy if you get a good deal, so not quite the same.
another one to try - some people would frown - is the LM 5532. a lot of high end recording gear has been using that opamp for years. for example, my tascam DA78hr high def 24bit digital tape recorder uses them. i find them to be nice as inputs because they are just open and clear - really open, really clear. the same thing would make them mildly unpleasant as outputs though. they're also really chep - about a buck each.
cheers and happy recording.
AT
Hey Adam
it's a Motu 2408 Mk3 PCI, not 828 firewire - no there are two of, and both will be modified! And Yes, PCB design is the same here as your picture shows, , but i have no mic-XLR inputs, only TRS
thats right, NJM4580 is the correct Name. As describrd in my last post, those 4 who are in between the (4) converters and belong to the output stage, i could see some connections wired to the sockets the output board is plugged in.
If you look to your attached Picture of the Main PCB Analog stage:
there are 12Pcs. 2115, and i think (haven't really followed the connections, but saw other gear doing same) "each triangle" of NJM 2115 's serve 2 channels, 1 OPAMP (per channel) is inverter (from symetrical to assymetrical) and from there one stage of an 2115 will buffer the signal to ADC.
(i'm not sure if it's clever to NOT change them....?)
Yeah, what u call LM5532 - is it same as well known and often used NE5532, but made from National? Or is there a difference in quality/design ?
it's a Motu 2408 Mk3 PCI, not 828 firewire - no there are two of, and both will be modified! And Yes, PCB design is the same here as your picture shows, , but i have no mic-XLR inputs, only TRS
thats right, NJM4580 is the correct Name. As describrd in my last post, those 4 who are in between the (4) converters and belong to the output stage, i could see some connections wired to the sockets the output board is plugged in.
If you look to your attached Picture of the Main PCB Analog stage:
there are 12Pcs. 2115, and i think (haven't really followed the connections, but saw other gear doing same) "each triangle" of NJM 2115 's serve 2 channels, 1 OPAMP (per channel) is inverter (from symetrical to assymetrical) and from there one stage of an 2115 will buffer the signal to ADC.
(i'm not sure if it's clever to NOT change them....?)
Yeah, what u call LM5532 - is it same as well known and often used NE5532, but made from National? Or is there a difference in quality/design ?
Hello all-
Just a little FYI, I have seen the NE5532 in a LOT of high-end recording gear over the years, it seems to be a fairly good quality unit that performs very well when properly powered, as well as laid-out correctly on the circuit board. It is often used in some very high-end (and high-cost) units. I just took a look at my PDF of the schematic copy I have here for the analog output card on our DiGiCo D5 console at work, and the circuit they are using has a NE5532 feeding a Analog Devices SSM2142 that's drive the balanced line output stage for the D-to-A cards. The DiGiCo is one of the higher-end consoles out there, it is used as the main mixing console for bands like U2 and Madonna, as well as Mark Knopfler and even Pavarotti. (I don't personally run the audio at those kind of events, mind you. That's for the guys above my pay grade. LOL!)
There are quite a few considerations on what affects the output quality of a given op-amp in a circuit, and if I had to make a estimation on the MOTU box it will more than likely be limited in the end by the quality of the power supply that it uses to feed the various op-amps. Here's a great discussion on op-amp power issues on this website: Tangentsoft - Audio
Also, here's a link to a fabulous discussion between Mike Stoica (pres. of Sytek Audio) and Jim Moss that comments on some of the common preamp designs in a few of the nicer analog mixing consoles over the years. They even "name names" on the various component model numbers of the ICs when doing it! It is fabulous reading for anyone interested in knowing some of the thoughts that a couple of well-respected guys in the audio recording world think makes a "musical sounding" unit perform as such:
Jim Moss/Mike Stoica Interview Link
FYI - Sytek Audio makes the well-regarded analog consoles in the Neotek line, I can personally attest to the excellent sound of both their "Elite" and "Elan" series units!
HTH-
Just a little FYI, I have seen the NE5532 in a LOT of high-end recording gear over the years, it seems to be a fairly good quality unit that performs very well when properly powered, as well as laid-out correctly on the circuit board. It is often used in some very high-end (and high-cost) units. I just took a look at my PDF of the schematic copy I have here for the analog output card on our DiGiCo D5 console at work, and the circuit they are using has a NE5532 feeding a Analog Devices SSM2142 that's drive the balanced line output stage for the D-to-A cards. The DiGiCo is one of the higher-end consoles out there, it is used as the main mixing console for bands like U2 and Madonna, as well as Mark Knopfler and even Pavarotti. (I don't personally run the audio at those kind of events, mind you. That's for the guys above my pay grade. LOL!)
There are quite a few considerations on what affects the output quality of a given op-amp in a circuit, and if I had to make a estimation on the MOTU box it will more than likely be limited in the end by the quality of the power supply that it uses to feed the various op-amps. Here's a great discussion on op-amp power issues on this website: Tangentsoft - Audio
Also, here's a link to a fabulous discussion between Mike Stoica (pres. of Sytek Audio) and Jim Moss that comments on some of the common preamp designs in a few of the nicer analog mixing consoles over the years. They even "name names" on the various component model numbers of the ICs when doing it! It is fabulous reading for anyone interested in knowing some of the thoughts that a couple of well-respected guys in the audio recording world think makes a "musical sounding" unit perform as such:
Jim Moss/Mike Stoica Interview Link
FYI - Sytek Audio makes the well-regarded analog consoles in the Neotek line, I can personally attest to the excellent sound of both their "Elite" and "Elan" series units!
HTH-
thanks for the thoughts guys.
yes, i meant the ne5532, just called it wrong.
i'll have to check out that article.
i'll also have to change those other 'third' opamps ; )
AT
yes, i meant the ne5532, just called it wrong.
i'll have to check out that article.
i'll also have to change those other 'third' opamps ; )
AT
to cliff45,
NE5532 is very popular and very often used in better audiocircuits, possibly best seller worldwide, they are really good - AND they are cheap. That makes them very interesting if u have to buy a lot of. You won't spend 2$ each opa if u can get nearly same qualitys for, say 27 pennys....
"this is the best", said from a Manufacturer means basically "for the money"
if i DIY change 30 Op's, better soundquality is goal, not the 25 bucks i could save...
MS
NE5532 is very popular and very often used in better audiocircuits, possibly best seller worldwide, they are really good - AND they are cheap. That makes them very interesting if u have to buy a lot of. You won't spend 2$ each opa if u can get nearly same qualitys for, say 27 pennys....
"this is the best", said from a Manufacturer means basically "for the money"
if i DIY change 30 Op's, better soundquality is goal, not the 25 bucks i could save...
MS
that being said, smitsch, i have in my possession two digital audio interfaces to record through - my modded 828 and a tascam da78hr (which i use the ADC's from basically, into my motu). which one do i like better and use more for getting good sounds? the tascam (with ne5532s in it OMG!). why? well, the motu sounds really a lot better than it did, and i prefer the ths4032 as my main outs because they're so sweet sounding. however, i actually prefer the less sweet sounding/more open sounding tascam. that also being said, the opamps are not the only factor in the equation at all. the tascam has: a far meatier power supply (my lights dim when i kicki it on!), it has different - perhaps better - ADC's in it, it has a different word clock system - runs on a typical crystal as opposed to the TI chip that motu uses.
the clock is the least of the differences i think, because when i had them both clocked to a big ben, i still preferred the tascam.
i have a handful of ths4032's set aside at the moment though. i am toying with whether i should try them in the tascam rather than finishing my motu.
if i think about it though, lots of amazing world class studios have made fantastic sounding records with the tascam digital tape recorders (the da 38 and 78 etc) and even the adat tape recorders. i reckon if it's good enough for them perhaps i shouldn't try to make it better... but that wouldn't be in keeping with being a DIYer would it?
all the best.
AT
the clock is the least of the differences i think, because when i had them both clocked to a big ben, i still preferred the tascam.
i have a handful of ths4032's set aside at the moment though. i am toying with whether i should try them in the tascam rather than finishing my motu.
if i think about it though, lots of amazing world class studios have made fantastic sounding records with the tascam digital tape recorders (the da 38 and 78 etc) and even the adat tape recorders. i reckon if it's good enough for them perhaps i shouldn't try to make it better... but that wouldn't be in keeping with being a DIYer would it?
all the best.
AT
- Status
- Not open for further replies.