I'm trying to get a handle on this I/V stuff, from what I understand I have in my Dac a PCM1738 d/a chip which is suppose to be a current output chip. Now from what I've been reading, is that a current output dac needs to see a low impedance load on it's output (input of the I/V stage) to keep it's distortion figures low, and it should be ideally a short to ground but practically 5-10ohms.
So why do manufacturers stick on a NE5532 or OPA627 or whatever opamps, they all have input impedances well over
1 megohm, nowhere near the ideal close to a short.
So this AD844 that I've posted is designed to be a proper I/V stage for a current output dac (if you read the data sheet down towards the end), with it's emitter loaded input I'm lead to belive that it is going to be very low input impedance, a so called prefect load for the dac? It should sound totaly different to what I've been playing around with once I get it.
Cheers George
So why do manufacturers stick on a NE5532 or OPA627 or whatever opamps, they all have input impedances well over
1 megohm, nowhere near the ideal close to a short.
So this AD844 that I've posted is designed to be a proper I/V stage for a current output dac (if you read the data sheet down towards the end), with it's emitter loaded input I'm lead to belive that it is going to be very low input impedance, a so called prefect load for the dac? It should sound totaly different to what I've been playing around with once I get it.
Cheers George
Anyone who has read an opamp textbook will know that in an IV amplifier, feedback brings the input (and output) impedances to vanishingly low levels, creating a "virtual ground" at the input.
Now, anyone who has actually understood an opamp textbook will know that said feedback effect is proportional to the open loop gain which is frequency-dependent ; therefore creating a "virtual ground" at 20 Hz and a very real load of bull at 1 MHz if you use a NE5532.
You can know in which category a CD player designer is by looking at which opamp he uses at the IV, heheh. The result is pretty scary if you ask me.
hint : NE5532 = 0.05 $ ; AD844 = 2.50 $
It is never a waste of time to write the feedback equations on a piece of paper
Check the datasheets, all of them mention the open loop gain versus frequency. Audio opamps tend to have a very high open loop gain at LF and a low open loop bandwidth. High speed opamps (video...) tend to have a lower OL gain, sustained up to much much higher frequency (up to several MHz for some high performance models).
Also, about the opamp you have in mind, I'd suggest you simply simulate the equivalent schematic in LTSpice or Microcap demo, and see what you get.
Now, anyone who has actually understood an opamp textbook will know that said feedback effect is proportional to the open loop gain which is frequency-dependent ; therefore creating a "virtual ground" at 20 Hz and a very real load of bull at 1 MHz if you use a NE5532.
You can know in which category a CD player designer is by looking at which opamp he uses at the IV, heheh. The result is pretty scary if you ask me.
hint : NE5532 = 0.05 $ ; AD844 = 2.50 $
It is never a waste of time to write the feedback equations on a piece of paper
Check the datasheets, all of them mention the open loop gain versus frequency. Audio opamps tend to have a very high open loop gain at LF and a low open loop bandwidth. High speed opamps (video...) tend to have a lower OL gain, sustained up to much much higher frequency (up to several MHz for some high performance models).
Also, about the opamp you have in mind, I'd suggest you simply simulate the equivalent schematic in LTSpice or Microcap demo, and see what you get.
I have build an I/V using AD844, Pedja Rogic's design. Use it with glue logic from I2S to PCM56 Nonos.
Strange thing is: on startup it sounds good, reasonable detailed (nonos!) nice wide soundstage. But after a longer period (warm up?) treble gets annoying bright. I use no filter yet. Soundstage is wide, but not very deep compared to dig. oversampled TDA1541.
Rightnow i am building a tube stage, and a passive I/V Res.
Strange thing is: on startup it sounds good, reasonable detailed (nonos!) nice wide soundstage. But after a longer period (warm up?) treble gets annoying bright. I use no filter yet. Soundstage is wide, but not very deep compared to dig. oversampled TDA1541.
Rightnow i am building a tube stage, and a passive I/V Res.
peufeu said:
Downloaded it once, and gave up in frustration trying to use it, I'm hopeless with software programs like that.
I just tried 2x AD812's which are dual current feeback opamps and they must of oscillated like crazy because they got stinking hot in a couple of seconds, and they are bypassed underneath at the sockett pins with 6.8uf tantalum and .1uf MTK at each + and - supply pins to ground. Hope the AD844 stays stable when I get it.
Cheers George
Just found this regarding current steering transimpedence amps for I/V use, some of you may understand it, I haven't read it yet and some will probally go over my head.
http://www.essex.ac.uk/ese/research...Current steering transimpedance amplifier.pdf
Cheers George
http://www.essex.ac.uk/ese/research...Current steering transimpedance amplifier.pdf
Cheers George
i/v output stage
Seems like a good place to get some advice here now, I am staring at my DV89's analog board and scratching my head, I want to mod the stereo pair, there are two WM8740's (dac) feeding differential to each a opa2134 one channel of in reverse polarity, the other channel is now an i/v output stage with a switchable gain selection (via analog monolithic switch shorting resistor to input resistor) from 1x to 2x (i guess for hdcd or something, so.....
if the IV stage is after the gain stage as in this case would a THS4032 be suitable candidate and if so... whom should I drive it with? (opa2132?? or lm4561??)??
Thanks for your time ;-)
any advice on mods for me greatly accepted!!
Seems like a good place to get some advice here now, I am staring at my DV89's analog board and scratching my head, I want to mod the stereo pair, there are two WM8740's (dac) feeding differential to each a opa2134 one channel of in reverse polarity, the other channel is now an i/v output stage with a switchable gain selection (via analog monolithic switch shorting resistor to input resistor) from 1x to 2x (i guess for hdcd or something, so.....
if the IV stage is after the gain stage as in this case would a THS4032 be suitable candidate and if so... whom should I drive it with? (opa2132?? or lm4561??)??
Thanks for your time ;-)
any advice on mods for me greatly accepted!!
Re: Re: i/v output stage
Dont think that they are readilly available my friend so I am flying with my eyes closed to some extent, I can understand most of the circuit, there is a filter built into the gain stage and the iv/out stage has sculpted decoupling and filtering (100ufdecoupling at that)
so what Im interested in is weather the lm4562 would be better, i was thinking of browndogging opa627/ths4031 but for the available technology in the unit a good supply rejection would probably be beneficial so the single channel of the lm4562 instead of the 627.
any way its based around a little something like this ...
tubee said:
Dont think that they are readilly available my friend so I am flying with my eyes closed to some extent, I can understand most of the circuit, there is a filter built into the gain stage and the iv/out stage has sculpted decoupling and filtering (100ufdecoupling at that)
so what Im interested in is weather the lm4562 would be better, i was thinking of browndogging opa627/ths4031 but for the available technology in the unit a good supply rejection would probably be beneficial so the single channel of the lm4562 instead of the 627.
any way its based around a little something like this ...
Attachments
Re: Re: Re: Re: i/v output stage
wish I could, but these are soic's and so I am looking for some direction before proceeding, I was wondering if anyone has experience using the named opamps with filters in the said configurations or similar, or even mods to ARCAM dacs as they seem to be of common design (albeit a little rough around the edges) by the way I meant to say 100uf coupling as in dc blocking..(I should change my username to TYPO or something) dc servo how to would be great...
Thanks for help!
tubee said:
wish I could, but these are soic's and so I am looking for some direction before proceeding, I was wondering if anyone has experience using the named opamps with filters in the said configurations or similar, or even mods to ARCAM dacs as they seem to be of common design (albeit a little rough around the edges) by the way I meant to say 100uf coupling as in dc blocking..(I should change my username to TYPO or something) dc servo how to would be great...
Thanks for help!
Has anyone tried LM6172? It is used for I/V conversion in Sony SCD-1 and SCD-777ES. One prblem is that it is a current feedback opamp. This means that some of the DAC output current has to go into the inverting input of the opamp, resulting in some error. The later model SCD-XA777ES uses opa2132s.
Re: Re: Re: Re: re: i/v output stage
So anyway I threw a pair lm4562 at it cos i'm impatient like that! and I must admit that I am impressed, I would like to see a comparison between settling at different capacitances between that and the 2134, tonally a lot more balanced its like someone has gone accross the sound stage with an iron! I found that the 2134 in this app was fatiguing in the upper mid, sometimes drowned out sometimes aggressive in a small bandwith. and so is it fast enough for the job well it sounds awight so Its awight, I like it.
So anyway I threw a pair lm4562 at it cos i'm impatient like that! and I must admit that I am impressed, I would like to see a comparison between settling at different capacitances between that and the 2134, tonally a lot more balanced its like someone has gone accross the sound stage with an iron! I found that the 2134 in this app was fatiguing in the upper mid, sometimes drowned out sometimes aggressive in a small bandwith. and so is it fast enough for the job well it sounds awight so Its awight, I like it.
ctong said:
I found it to be very good across the baud, very accurate and honest sounding, you know when you read posts when people say they hear things that they didnt previously I guess thats a colouration thats been wiped clean, like all music sounds different rather than having a particular tonal personality while at the same time I did'nt find it sterile and was able to listen to it for hours, I guess that the I/V capability under the curcumstances is adequate, It would be interesting to hear about some statistics as to what exactly is acceptable and what is overkill for audio and why,
The test data that NS supply shows for 100pf load which apparently isnt very audio and so would be much faster in an audio application with a less reactive load.
I dont know about that really can anyone back it up?
something else is that the psu in the DV89 is a bit
so the psrr of that opamp may well be beneficial in this instance.tubee said:
I installed DIP8 sockets on my the IV stage on my Sony SCD-C555ES to swap opamps. Unfortunately, THS4032 does not offer DIP8 packaging, making comparisons difficult. I might try LM6172, which is used in Sony SCD-1 and is said to be very revealing (=analytical and lean bass?) Anyone has experience with this opamp?
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.