Best resistors For I/V Conversion?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
I'm a big fan of wire wound resistors for precision applications, but there's no such thing as a zero inductance wire wound resistor, and some winding techniques also produce significant stray capacitance. I'd be a bit leery of using them for high bandwidth I/V conversion. The bulk metal foil resistors best wirewounds in just about every way except maybe peak power handling, and I'm not even sure about that. A decent metal film like an RN55 or 60 type is also no slouch. The value of the resistor would affect my choice of type. For low values the wire wound can be excellent. For mid values the bulk metal foil or MF wins, and for high values the MF is the logical choice. Don't go near thick film, carbon film, carbon, or MOX resistors due to instability, noise and non-linearity with voltage, in spite of sometimes excellent HF performance.
 
Anything under a couple hundred ohms is low to me, as I can wind it with a reasonable size (i.e. easy to handle) wire that comes out to a reasonable length. That means long enough to adjust to high accuracy, and short enough so the resistor is of reasonable size. I wouldn't dream of winding anything over a couple k at most, and that only for instrumentation and reference use, not audio. I wind quite a few of my own resistors and shunts, and use either manganin or an 800 ohm-cm wire of very low TC. The manganin is traditional and its properties are extremely well known. It can be easily soldered. The 800 ohm-cm wire cannot be soldered, but has to be spot welded. It has a lower TC than manganin. It's what you'd find inside many commercial wire wound resistors. I have some Mills resistors and think highly of them, as well as of Julie Research resistors, if you can find them. Still, bulk metal foil is really the high end resistor these days.
 
I found Siemens non inductive wirewound resistors to try in my DAC as passive I/V, value is 43,759 ohm. Looks like a relay coil and shines like silver when gold colored insulation is removed. 1,9m total wire length of 0,25mm wire have 43 ohm. There are additional windings with thinner copper wire for exact value tuning.
 
MOX wouldn't be my first choice for anything that needed an ideal resistor, based on voltage coefficient. I used some in the feedback path of a high voltage amplifier, and the non-linearity was easy to see on the scope. OTOH, I've never used or measured values that low, so maybe it's not an issue. IMO, they're cheap enough that you shouldn't trust anybody, but buy some and test 'em. For those values, a short piece or a counter wound coil of manganin wire would be nearly ideal.
 
Hello to all!

Does anyone have comments as to preferred resistor
types for digital I/V conversion, and the sound character
of any mentioned types/brands?

Thanks!

fastcat

TX2575. Very natural, able to deliver the speed, with no coloration of any kind. It sounds outstanding and it gives least amount of ringing and overshoots/undershoots. They also come in 0.1% tolerance, so matching exercise is usually not required.
 
I compared the Econistor Rhopoint you mentioned, with TX2575. The TX2575 was much better, at least in this (I/V) application. The problem with Econistor Rhopoint is related to the fact that they are wirewound - hence, there's inductance. You may think that the inductance is negligible, but it is in fact substantial to cause issues…. For example, if Econistor Rhopoint are used where inductance is absolutely undesirable (an example would be a feedback resistor in very fast, current - feedback OP's, which sound amazing as I/V if the gain is necessary - like with 1704's) the introduced inductance will take these OP's into oscillations and eventually would cause an OP damage, and destruction.

A big contributing factor to what sounds good as an I/V resistor is rest of the sound reproduction chain… The TX2575 may sound too revealing and the Econistor Rhopoint might be just right. I just provided my experience and findings for others to consider…
 
I compared the Econistor Rhopoint you mentioned, with TX2575. The TX2575 was much better, at least in this (I/V) application. The problem with Econistor Rhopoint is related to the fact that they are wirewound - hence, there's inductance. You may think that the inductance is negligible, but it is in fact substantial to cause issues…. For example, if Econistor Rhopoint are used where inductance is absolutely undesirable (an example would be a feedback resistor in very fast, current - feedback OP's, which sound amazing as I/V if the gain is necessary - like with 1704's) the introduced inductance will take these OP's into oscillations and eventually would cause an OP damage, and destruction.

A big contributing factor to what sounds good as an I/V resistor is rest of the sound reproduction chain… The TX2575 may sound too revealing and the Econistor Rhopoint might be just right. I just provided my experience and findings for others to consider…

Of course I agree, all is in the context : mine is just on a TDA1541A, a relative high value and bypassed with tinned foil styren cap ! :D So here despite the stray inductance of the wirewound it sound smoother and as transparent than a naked Ti Bulk Foil Var resistor ! At least as you say, with the whole Hi fi I have ! I remembered the father of the ESL63 didn't liked the TX with the TDA1543 for instance... so !
 
Of course I agree, all is in the context : mine is just on a TDA1541A, a relative high value and bypassed with tinned foil styren cap ! :D So here despite the stray inductance of the wirewound it sound smoother and as transparent than a naked Ti Bulk Foil Var resistor ! At least as you say, with the whole Hi fi I have ! I remembered the father of the ESL63 didn't liked the TX with the TDA1543 for instance... so !

Exactly right! On the side note, and in an attempt to make them look good, they are (the Rhopoint's) encapsulated in plastics that could also bring sound benefits...

Regards,
Nick
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.