Quad ESL57 input transformer ratios

[Imazed]

I am attempting a rebuild of a couple of ESL57. Does anyone know the transformer ratios or test voltages. I get different results from the two I have and I do not know what is correct.

 

[stokessd]

I measured the ratios years ago, and this is what I came up with. It could be off by a turn or two:

Treble: 1:96
Bass (effective; includes treble): 1:270

Sheldon
quadesl.com

 

[bolserst]

I have not personally measured ESL57 transformers, but the transformer design is discussed by Peter Baxandall on pages 169 – 179 of “Loudspeaker and Headphone Handbook”, J. Borwick, Ed. The ratios stated were 1:87 and 1:290…pretty close to what stokessd mentioned. Further details on the transformer parasitics and filtering of the different panels is also discussed in detail in the book. I short summary was posted here: Explaining the Quad ESL-57 Crossover

FWIW, my measurements of the ESL63 transformers match well with the data provided by Baxandall in the same book, offering some level of confidence in the ESL57 numbers.

 

[Imazed]

Thank you Sheldon. That was very useful as I had several ESLl57 giving different (and varying) output for the same input. I eventually stripped off all crossover components and measured the three secondary windings. One (pins 3,7) has a ratio of 1:99, is centre tapped to earth (pin 16) and drives the treble units. The two other windings(Pins 1,2 and 4,5) have a 1:95 ratio. All three windings are connected in series to drive the bass units so the bass ratio is 1:289 Readings are averages of three units.

I then added new crossover components to one unit and got the following voltage ratios Mid Treble (blue wires) 1:96 High Treble (brown wires) 1:50 and Bass 1:289

 

[Imazed]

Thank you boiserst. Lovely to know there is definitive information somewhere and that my measurements were not too far out. I have been meaning to get a copy of “Loudspeaker and Headphone Handbook” for a while. That clinched it so waiting patiently for the postman now

 

[lcsaszar]

Could it be that the original intention was 3 x 1:100 windings, and the actual measurements come from manufacturing tolerances?

 

[Imazed]

Could it be that the original intention was 3 x 1:100 windings, and the actual measurements come from manufacturing tolerances?

I doubt it. I am certain that my measurements are a) limited by equipment and skill and b) carried out on actual transformers which are subject to manufacturing tolerance.
I am equally certain that the ratios quoted in “Loudspeaker and Headphone Handbook” , which I now have, are correct because they match with information I have found in other old articles.



Also I am now in possession of a few more transformers which measure (on average) 3 windings of 96.5:1. Furthermore the secondary windings have around 10,000 turns so to have a manufacturing change from 100:1 to 97:1 would mean the automatic winding machines miscounted by 300 turns - very unlikely.



Regarding manufacturing tolerances the widest difference I found on six transformers was less than 1:0.4. so I guess my 96.5 would be within manufacturing tolerance of 97:1.

The other reason why I have a slightly different ratio is that I tested using standard un-calibrated multimeter at low frequency