Practical advice for zero-field transformer use?

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

mikep

Well-known member
Joined
Feb 18, 2006
Messages
450
Location
Philadelphia
Im trying to use ordinary mic and line transformers in zero-field configurations and getting less than stellar results. it works but the performance is a little worse than using the transformer as they were intended. I think it is mostly an issue of me using completly the wrong parts. I have some ideas, but does anyone know the basics of selection criteria for this application? I know the core doesn't have to be as large, but it doesn't hurt if it is, right? what about primary inductance and trurns ratios?
 
We need more information: show us a schematic (there are several ways to implement this) and give us the transformer types you are currently using.

As a first answer: secondary resistance is very important (unless you work with negative impedances in the active part), as this limits how close we get to zero.

I guess after searching you've already found this thread: www.groupdiy.com/index.php?topic=10441

Samuel
 
Thanks samuel. Yeah I saw that thread. as for a schematic, I think I should play this one close to the chest until I figure out if it is going to work. it is not a standard configuration.

[quote author="Samuel Groner"]
As a first answer: secondary resistance is very important (unless you work with negative impedances in the active part), as this limits how close we get to zero.l[/quote]

hmm. maybe this is what I am doing "wrong"? I am trying to do something at lower-Z but not zero. does it have to be close to zero to get good performance? Can't imagine why. I am doing it partly to increase the headroom of a small input transformer. it works, but the distortion is a little higher. with one transformer, a 10k:10k sowter line input, the measured performance IS improved but the subjective sound quality is still worse. various higher ratio parts (lundahl, shure and beyer 1:5, 1:7, 1:10) are worse still. the degredation seems to be related to how much series resistance I put at the front end.

I just wonder what kind of part is commonly used in a zero-field circuit. are they usually 1:1 ? I was looking at the lundahl zero-field line input datasheet, there are few specs, no mention of ratio, etc.

has anyone looked inside the new tubetech summing box? it supposedly has Z.f. inputs.
 
I was looking at the lundahl zero-field line input datasheet, there are few specs, no mention of ratio, etc.
Look again, they specify turns ratio.

Does it have to be close to zero to get good performance?
Yes. A standard line input (you mentioned a 10k:10k) is about the worst thing you can get--everything above a few dozend ohms secondary DC resistance will not work unless you use negative impedances to cancel things.

Samuel
 
mikep said:
I just wonder what kind of part is commonly used in a zero-field circuit.  are they usually 1:1 ?  I was looking at the lundahl zero-field line input datasheet, there are few specs, no mention of ratio, etc.

The little transformers in the NTP ZFT modules are reported to be 1:3   (135 Ohms : 1450 Ohms)   see patent 5130662.

Apart from the secondary resistance I'm wondering as well if there are any other special requirements for TXs to be suited for
zero-field applications...
would for instance be nice to be able to convert 'substandard TX-stages' into decently behaving inputs.

Regards,

 Peter
 
mikep said:
hmm.  maybe this is what I am doing "wrong"?   I am trying to do something at lower-Z but not zero.   does it have to be close to zero to get good performance?  Can't imagine why.  I am doing it partly to increase the headroom of a small input transformer.  it works, but the distortion is a little higher.  with one transformer, a 10k:10k sowter line input, the measured performance IS improved but the subjective sound quality is still worse.  various higher ratio parts (lundahl, shure and beyer 1:5, 1:7, 1:10) are worse still.  the degredation seems to be related to how much series resistance I put at the front end.

..I'm just a hack, thinking out loud, so you might want to ignore this post.. ;D
isnt "Zero Field" just a fancy way of saying current sensing input?..if so, that would explain why they allways use the "-" (current sensing) input on the opamp. ...and a 1:10 voltage ratio transformer is 10:1 current ratio..perhaps the transformer should be worked backwards?... ???
..again, just a hack, thinking out loud...
j
 
I have used zero-field inputs in many units; I used a standard Sowter mic xfmr rated at 600:600, 2x 4.7k resistors at the input, with 4.7k/100pF RC's around the opamp. I never had any issue with that.
 
abbey road d enfer said:
I have used zero-field inputs in many units; I used a standard Sowter mic xfmr rated at 600:600, 2x 4.7k resistors at the input, with 4.7k/100pF RC's around the opamp. I never had any issue with that.


Thanks for the info, sounds good. The idea was to ZF-convert a box with lots of small iron into one that makes a better chance of coping with line-levels (pics here http://www.groupdiy.com/index.php?topic=35838.msg440076#msg440076 )

The thing is (was), certain TXs are mentioned/specified for zero-field use, so that makes one wonder...

but then again, some mics are specified for certain sources, but that doesn't mean that... etc

Regards,

 Peter
 
Optimising a transformer for zero-field makes it an odd beast that can hardly be used for any other standard application, so it has to be identified as such, but the use of a reasonably sized standard xfmr compromises very little in performance and makes purchasing way easier.
 
This makes sense !


Makes one also wonder why ZF is relatively little used... saves cost, weight, resources... you need a subsequent amp anyway...


Regards,

  Peter
 
I guess most of the time you would need that amp anyway. I suspect many designers are afraid of transformers because they haven't had a formal training on them. I found out, talking to students, that their exposure to transformers during their electronic courses is limited to: expensive, bulky, non-linear, stay away from them...
I must say that, when everything is said and done, the only "real-world" advantage of xfmrs compared to active designs is galvanic isolation, and it takes some time to optimise a xfmr based input or output stage.
 
Hi,

It's indeed from where you approach. The comparison I made in my previous message
was between the usual bigger input TXs vs using a smaller probably cheaper one in ZF-config.

But there'll be indeed fear / unknownness (a word?) for using a TX alltogether, that's the other approach.

Didn't do the exact math, but ZF will be less easy to do with tubes I expect... at least require additional loopgain on top of what the stages would need to provide when been used with a 'normal' input-TX.

Obviously a solid state opamp does enough open-loop gain for making a decent virtual earth.

Enough rambling from my side for now; the exact explanation of relatively scarce ZF-use might probably be unknownness(..), historical reasons & a final BOM-calculation that perhaps doesn't show the expected savings.

Regards,

  Peter
 
abbey road d enfer said:
I have used zero-field inputs in many units; I used a standard Sowter mic xfmr rated at 600:600, 2x 4.7k resistors at the input, with 4.7k/100pF RC's around the opamp. I never had any issue with that.

I'll have another look at the NTP-patent to learn how much the use of a second opamp really adds. You seem to be doing fine with just one.
 

Latest posts

Back
Top