Ring Stinger transformer substitution?

[anotherjim]

Actually, my Eagle traffy has another part number - P631M. Blue wrap and 20k:1k+1k. Made in Taiwan.
20k primary DCR is 1k and the secondary total is 100R.

[Govmnt_Lacky]

The plain old multimeter ohm test isn't detecting anything either, and since I'm surprised by this, clearly I need to learn a thing or two about how transformers work.

Hmmmm.....

I have only ever used the Blue ribbon LT44s however, I have gotten a resistance reading from each side of the transformer every time (except once when I received one with a busted primary connection between the lead and the winding)
Unless your DMM has a higher-than-usual 'low resistance threshold' then you might want to double check to make sure you didn't get some bunk parts 

[digi2t]

I only have, and ever used, blue ones. I have some in the shop, so I took a measure.

Resistance on the primary, 1.2K.
Resistance on the secondary (outer pins), 110 ohms.

With my inductance meter;
Primary, 15.3H.
Secondary (outer pins), 791mH.
Secondary (center tap to an outer pin), 131mH.

[anotherjim]

I think auto-ranging meters might take a while to resolve a coil DCR. I watched mine messing about until it settled after what seemed a suspiciously long time. Maybe if it switches test current, the reactance fools it?

[aion]

It may have just been corrosion or wax on the transformer leads. I roughed them up a bit and got resistance readings with the DMM. This morning I didn't get a single reading on any of 5 of them (including a Xicon one as a control), so I concluded that my methodology was wrong somehow, but nope... can't get very far without electrical contact. Here are mine.

Blue #1
Primary: 1.170k
Secondary: 108R 1-3, 55R 1-2. 55R 2-3

Blue #2
Primary: 1.184k
Secondary: 108R 1-3, 55R 1-2, 53R 2-3

Yellow #1
Primary: 490R
Secondary: 109R 1-3, 55R 1-2, 54R 2-3

Yellow #2
Primary: 489R
Secondary: 111R 1-3, 57R 1-2, 56R 2-3

So based on that, I wonder if the yellow one is 10k primary while the blue is 20k primary. That would explain where the spec discrepancy comes from since both are called LT44. The secondaries are identical for both, in resistance at least.

[Rob Strand]

Here's some analysis of the measurements.

To make any further conclusions tests on primary and secondary voltages
at 1kHz (and technically correct terminations) would be required.

--------------------------------------------------------------------------------------------------
LP44 Transformer (Blue,  PM-631M)                       
Rob 30 Dec 2021                       
                       
1) Estimate turns ratio from inductances                       
                       
Based on digit2t's Blue Unit Inductances (29 Dec 2021, reply #22)                       
Lp [H]        15.3               
Ls [H]        0.791        Outside       
Ls1 [H]       0.131        To Center       
4*Ls1 [H]     0.524               
                       
These measurements can only be trusted so far as the winding                       
capacitance can significantly change the inductance measurement.                       
                       
Ideally Ls and 4*Ls1 should be equal but we have                         
variation due to the core drive level or capacitance                       
                       
The turns ratio n= Np/Ns can be from sqrt(Lp/Ls)                       
(Np/Ns) low       4.40               
(Np/Ns) hi        5.40               
                       
Assuming a Zp we can estimate Zs = Zp/n^2                       
Zp [ohm]               20000                **** Assumed value
Zs est [ohm] hi        1034.0               
Zs est [ohm] low       685.0               
                       
So a secondary impedance of 1k is consistent with a 20k primary impedance.                       
                       
2) Estimate turns ratio from DC resistance                       
Based on aion's resistance measurement (29 dec reply #24)                       
we can also makes some crude estimate of the turns ratio                       
n  = Np/Ns = sqrt(m) * sqrt(Rp / Rs)                       
where m is the mean length turn (MLT) ratio of the primary and secondary                       
If the secondary is wound over the primary m = 1.5 (approx)                       
                       
Rp [ohm]        1170               
Rs [ohm]        108        outside         
m               1.5               
                       
n = (Np/Ns)        4.03               
                       
Not bad agreement with  the inductance based estimate                       
                       
3)  Impedance to DC resistance checks                       
                       
Zp/Rp        17.1               
Zs/Rs        9.6               
                       
We expect values around 10 to 20, higher for primary if that is wound first.                       
                       
4) Conclusion                       
                       
Blue LT44 (PT631M):                       
The measurements are consistent with a 20k to 1k transformer.                       
Where 1k is to the outsides of the secondary.                       
                       
This agrees with the crude data supplied with the transformer.                       
                       
Yellow LT44:                       
The primary resistance of 490 ohm which is half the Blue transformer.                       
The secondary resistance is similar to the blue transformer.                         
It seems reasonable to assume this is a 10k to 1k transformer.                       
                       
This agrees with the data given on the Velleman website.                       

[digi2t]

This may support that the yellow LT44 is indeed 10K:1K...

https://www.esr.co.uk/shop/contents/en-uk/p33681_Audio_Output_Transformer,_Pri_10k,_Sec_2_x_1k_(LT-44).html

[Rob Strand]

This may support that the yellow LT44 is indeed 10K:1K...

https://www.esr.co.uk/shop/contents/en-uk/p33681_Audio_Output_Transformer,_Pri_10k,_Sec_2_x_1k_(LT-44).html

The problem with that is they write 2x1k.  It's like something has gone wrong in the translation.

I think part of the problem is the way the Eagle brochure writes 1k and CT amongst the windings.   It's easy for someone to misinterpret that as 1k for one of the windings.   However, I believe the correct interpretation is 1k for the outside then that winding has a center-tap.  (The tables for the Xicon transformers are written with the impedance for the outsides of the center-tapped winding.)

If you look here they are much clearer about it.
https://www.velleman.eu/products/view/?country=nl&lang=en&id=11604

In fact they give the turns
Np = 2500
Ns = 400 x 2  = 800 to outside
n = Np/Ns = 2500/(400*2) = 3.125

If we assume Zp = 10k then Zs = 10k/3.125^2 =  1024 ohms

So this is consistent with a 10k to 1k transformer.  As derived, and how I read the specs, 1k is to the outsides.
The winding is center-tapped but the 1k is to the outside.

[digi2t]

I believe the 2 x 1k to be a misnomer as well. I saw the yellow xfo elsewhere as well (Velleman I think), and it was listed as 10k:1k.

Regardless, I've never tried the yellow xfo in a RS build, and the gutshots I've seen of the original show blue tape xfos mounted to the baby board.

[aion]

Very interesting. So if it's 20k:1kCT, with 1k being the outside impedance, then the 42TM006 is a direct substitute.

...And of course, after all this, the 42TM006 is one of the few in the series that is not regularly stocked - not just out of stock, but not stocked at all. I'll send Mouser an email and see if they'd consider carrying it.

But in the mean time, I suppose we're back at the question of substitutes... I see that the transformers are symmetrical, secondary to secondary on the inside with the primaries on the outside, so it would seem like the second transformer reverses the step-down of the first one. Are the impedance and the ratio of the transformers themselves important, or only the P:S:S:P symmetry? And is a step-down even required? As mentioned in the original post, all other ring mods I've seen seem to use 1:1 transformers (10k:10k or 3k:3k). There's of course a lot more going on in the Ring Stinger, including the carrier/octave mode switching, but curious whether the overarching application is the same as other ring mods or whether the transformer impedance and ratio are specifically important here where they wouldn't be in simpler circuits.

[Rob Strand]

Very interesting. So if it's 20k:1kCT, with 1k being the outside impedance, then the 42TM006 is a direct substitute.

...And of course, after all this, the 42TM006 is one of the few in the series that is not regularly stocked - not just out of stock, but not stocked at all. I'll send Mouser an email and see if they'd consider carrying it.

If we think about the turns ratio then the 42TM019-RC might be an option. 

We have,
42TM006-RC 20k to 1k,  Rp=900 ohm, Rs=70,  n = 5:1
vs
42TM019-RC 10k to 600 ohm, Rp=600 ohm, Rs=100, n=3.77:1

So we are talking 30% difference in the voltage.  Not an enormous difference.

The 10k vs 20k might not be a big deal either since the impedances driving the transformer are quite low.

The best thing would be to try it and see how much difference, if any, is noticed.   Worst come to worse perhaps some circuit tweaks are possible to improve the match between the *sound of* the old and "new" transformer.   For example R10, R12, R13, R60?.

[Rob Strand]

Assuming the 10k primary is OK we can go the other way with the turns ratio.
42TM002     Impedance 10k to 2k, Resistance Rp=600ohm & Rs=155 ohm, turns ratio n=2.357:1.

Not as close as the 42TM019 but it might turn out having a lower rather than higher turns
ratio is easier to deal with.  Just an idea.

Perhaps even pitting the Yellow (LT44 10k 1k ct) and Blue (LT44/P631M 20k:1k ct) Eagle transformers against each other will tell you which way is better.   (I'm not even sure which Eagle transformer was used in you project originally.)

[aion]

(I'm not even sure which Eagle transformer was used in you project originally.)

The Ring Stinger definitely used the blue one based on the photos:



This does assume blue and yellow color coding has always been consistent. Newer ones don't have LT44 stamped on the top, so there have at least been cosmetic changes along the way, and Eagle isn't strong in the area of documentation. But folks have used current-production blues and compared their DIY builds against real Ring Stingers with success so I think it's as close to confirmed as we'll get.

[Rob Strand]

This does assume blue and yellow color coding has always been consistent. Newer ones don't have LT44 stamped on the top, so there have at least been cosmetic changes along the way, and Eagle isn't strong in the area of documentation. But folks have used current-production blues and compared their DIY builds against real Ring Stingers with success so I think it's as close to confirmed as we'll get.

Ah, ok.    Since the blues are 20k to 1k  that means the turns ratio is n=SQRT(20000/1000)= 4.472:1; or 5:1 according to the xicon 42TM006-RC.    The 42TM019 (n=3.77:1) is closer as far as turns ratio is concerned.  The 42TM002 (n=2.357:1) would be going too far away from the blue LT44/P631M.

The change from 10k to 20k input still needs to confirmed.   However, the 42TM series almost forces the 10k models due to availability.

I suspect there could minor changes in the LT44 specs over time.  Perhaps not so much the impedances but the DC resistances.  After scrounging through a few posts it seems some of the reported DC resistances didn't match either the yellow or blue values you measured values or the info on the ebay advert.  While it wasn't clear if the posters had the yellow or blue the fact it didn't match any means something has varied.

No doubt similar "equivalent" transformers were made for radios over time and probably all had minor variations.  I found reference to the LT44 in a 1973 electronics magazine.   The magazine reported the Eagle LT44 having a DC resistance less than 1k but a similar Rex LT44 having a DC resistance around 2k ohm.   Clearly at that stage the specs were still open.
[The Rex LT44 goes back to1965, Practical Electronics Feb1965!]

(FWIW,  there's also a red transformer the LT700.   The color coding of that seems consistent as well.  The specs are quite different to the LT44's.)

It looks like the Rex LT44 started life as a 20k to 1k ct transformer (n=4.472:1).   The primary DC resistance of 2k is OK for a 20K impedance transformer.  Clearly the later blue Eagle LT44 with the same impedance specs but lower primary DC resistance is a better wheel, at least in terms of losses.

This is from Henry's radio catalog 1964/1965,

[anotherjim]

This thread is probably the only place where the blue or yellow tape is explained. Elsewhere the opinion is that it was nothing more than a cosmetic change. I must admit I was suspicious since the LT700 output version was never anything but red taped. I've never seen an Eagle transformer with anything printed on it so logically, the tape colour has to be the means to identify its type.

[digi2t]

This thread is probably the only place where the blue or yellow tape is explained. Elsewhere the opinion is that it was nothing more than a cosmetic change. I must admit I was suspicious since the LT700 output version was never anything but red taped. I've never seen an Eagle transformer with anything printed on it so logically, the tape colour has to be the means to identify its type.

I must admit, when I built my String Ringer way back when, I faced the same dilemma, yellow or blue? The only thing that pushed me to use the blue xfos was the fact that the original did. I think one thing that almost derailed my choice on this was the UK Ebay seller stating that the yellow one was better geared for ring mods. Ultimately, I went with the blue ones anyway, and they worked fine.

But... it sure would be nice to know if one or the other would make any difference here, as well as what can be used as a substitute. Like Jim says, I've never come across a definitive source of information on this.

[Rob Strand]

I think one thing that almost derailed my choice on this was the UK Ebay seller stating that the yellow one was better geared for ring mods. Ultimately, I went with the blue ones anyway, and they worked fine.

The funny thing is I found a site which used the blue ones for a ring mod and it pointed to the ebay advert.   At the time I thought the page was written by the seller!