Partridge TG 6556/2 Output Transformer
- Thread starter CJ
- Start date
Help Support GroupDIY Audio Forum:
what is going on here?
simple. one green wire is a speaker lead, the other loops in and out of the bobbin (see early pics) as to patch in to another wind ,
maybe over thinking here, there are 3 sec winds but two leads, one lead hooks up with the first two secs which get better linkage because they are closer to the core, the outer wind gets it's own lead as to lower lead resistance which can be important when you are talking about a 0.25 ohm speaker wind, so the two leads might equalize current flow by offering less resistance to the least linked wind,
simple. one green wire is a speaker lead, the other loops in and out of the bobbin (see early pics) as to patch in to another wind ,
maybe over thinking here, there are 3 sec winds but two leads, one lead hooks up with the first two secs which get better linkage because they are closer to the core, the outer wind gets it's own lead as to lower lead resistance which can be important when you are talking about a 0.25 ohm speaker wind, so the two leads might equalize current flow by offering less resistance to the least linked wind,
Attachments
and that's a wrap.
some people have attached significance of the superior performance of certain insulating materials, dialectic constants and all that, since this xfmr was done in the paper bobbin days, they probably wanted an upgrade, kind of like Neve using those phenolic bobbins in the green meanie inputs,
ok so we have a coil structure of P-S-P-S-P-S-P done on a EI 125 lam with 1.5" stk ht,
373 turns per pri = 1492 three sec winds put in parallel which have 68 turns for the 8 ohm spk and an additional 29 turns to make 97 turns for the 16 ohm tap,
so the PRR method was pretty close, this is a 3.8 K to 8/16 ohm job with .006" lams for better bass and possibly better mid-range due to the increased freq response of the lams due to less eddy current,s got a print almost done,
some people have attached significance of the superior performance of certain insulating materials, dialectic constants and all that, since this xfmr was done in the paper bobbin days, they probably wanted an upgrade, kind of like Neve using those phenolic bobbins in the green meanie inputs,
ok so we have a coil structure of P-S-P-S-P-S-P done on a EI 125 lam with 1.5" stk ht,
373 turns per pri = 1492 three sec winds put in parallel which have 68 turns for the 8 ohm spk and an additional 29 turns to make 97 turns for the 16 ohm tap,
so the PRR method was pretty close, this is a 3.8 K to 8/16 ohm job with .006" lams for better bass and possibly better mid-range due to the increased freq response of the lams due to less eddy current,s got a print almost done,
Attachments
The Partridge were reputed to be some of the best , it definately seemed to give Hiwatt an edge in terms of volume and low distortion compared to Marshall .
Was it only a partial rewind was needed on that , or will you re-do it from scratch later on ?
Was it only a partial rewind was needed on that , or will you re-do it from scratch later on ?
Top stuff.
What would you indicate was the voltage withstand rating from Pri-Sec - as in the lowest creepage, or likely multi-layer solid insulation rating, or other weak point such as lead-outs ?
Did you do a turns ratio voltage test at the start, or was the unwound turns used for the impedance ratio, or was that in a partridge datasheet, or just the generic values used by people ?
What would you indicate was the voltage withstand rating from Pri-Sec - as in the lowest creepage, or likely multi-layer solid insulation rating, or other weak point such as lead-outs ?
Did you do a turns ratio voltage test at the start, or was the unwound turns used for the impedance ratio, or was that in a partridge datasheet, or just the generic values used by people ?
i can see why there are not a lot of these transformers left, this is the first power output i have seen with no poly insulation added to the paper insulation. add in aging and lack of varnish to all layers and you can bet this thing is ready to blow any second,
Doh, just saw the other post, and the image with the winding details in this thread!
Was the PRR test to wind on a known number of turns - 20T and energise another winding, and then measure the voltage on non-energised windings? And then swap to another energised winding, to avoid error in the turns ratio?
Was the PRR test to wind on a known number of turns - 20T and energise another winding, and then measure the voltage on non-energised windings? And then swap to another energised winding, to avoid error in the turns ratio?
here is a Build chart for the Partridge xfmr.
you will find these in commercial shops.
when an engineer designs a transformer, he adds up the thickness of the wire and insulation as he goes along to make sure everything fits. he knows how much space of say a piece of 10 PB ( 10 mill press board)will really take up when it is taped into place, this will obviously be a bit over the actual thickness of the press board, as it will not lay perfectly flat, there will be small deviations in the thickness from roll to roll, tape must be used to hold it into place, and some winders are stronger than others.
wire also has a +/- tolerance and the amount of tension will determine how much the coil grows with each layer. So an experienced engineer can usually get a design right the first time,
So a build table helps the winder and the engineer make it work out. If the first build is successful, then the winder will no longer need to put a pair of calipers over the coil each time he winds that model.
so what i did was measure the Partridge coil as it was taken down, so that we might a better shot of keeping the scrap bin empty of expensive copper and also save some time over having to wind the coil all over again, as if the coil is building out quicker then this table, we need to back down on wire gauge, use more tension, or change insulation and use less tape.
so all those engineers do is play with cardboard , wire and tape, 4 years of college and they give you a pair of scissors and a roll of wire, and put you on the day shift, lookout,
this is a rough guide that has not been proven yet but it is better than nothing,>
you will find these in commercial shops.
when an engineer designs a transformer, he adds up the thickness of the wire and insulation as he goes along to make sure everything fits. he knows how much space of say a piece of 10 PB ( 10 mill press board)will really take up when it is taped into place, this will obviously be a bit over the actual thickness of the press board, as it will not lay perfectly flat, there will be small deviations in the thickness from roll to roll, tape must be used to hold it into place, and some winders are stronger than others.
wire also has a +/- tolerance and the amount of tension will determine how much the coil grows with each layer. So an experienced engineer can usually get a design right the first time,
So a build table helps the winder and the engineer make it work out. If the first build is successful, then the winder will no longer need to put a pair of calipers over the coil each time he winds that model.
so what i did was measure the Partridge coil as it was taken down, so that we might a better shot of keeping the scrap bin empty of expensive copper and also save some time over having to wind the coil all over again, as if the coil is building out quicker then this table, we need to back down on wire gauge, use more tension, or change insulation and use less tape.
so all those engineers do is play with cardboard , wire and tape, 4 years of college and they give you a pair of scissors and a roll of wire, and put you on the day shift, lookout,
this is a rough guide that has not been proven yet but it is better than nothing,>
Attachments
Your right! Thanks Volker! I will fix soon...
clintrubber
Well-known member
CJ said:
Just curious how close the various 'Hiwatt Style TXs' actually are.
Which ones did you happen to see ?
Here's for instance a TubeTown DR103 (100W) Output Transformer, which I have in a DR103 replica:
https://www.tube-town.net/info/datenblaetter/transformatoren/tt-dr103-ot.pdf
(still working, but if it breaks I'll send it ;-) )
Bye
nobody is using the .006" lams, or the plastic bobbin, margins are different on a nylon bobbin (winding length is longer) probably different insulation, different processing (varnish vs air dry epoxy) probably still sounds good but not the same,
clintrubber
Well-known member
CJ said:
I see, could understand, too many differences for not having an influence. OTOH, for an as-HiFi-sounding-TX as these relatively are in guitar-amp-land, those differences might be possibly a moot point...?
Either way, fascinating dissection, thanks
, and as I understood it actually a bit surprisingly some corners were cut with those tap connections, as you showed. The late Mr Joyce might not have approved... :
we are going to do this rewind like the original so the customer does not lose resale value with a different looking transformer,
looks like the #28 primary wire is coming in with a heavy build so we are bumping it down to #29 ,
that looks like a nice transformer, they did the core bracket real nice, and those terminals make more sense,
i did see a set of Partridge transformers that came in on a Sound City amp, they looked just like this hi-watt
new print with the #29 AWG>
looks like the #28 primary wire is coming in with a heavy build so we are bumping it down to #29 ,
that looks like a nice transformer, they did the core bracket real nice, and those terminals make more sense,
i did see a set of Partridge transformers that came in on a Sound City amp, they looked just like this hi-watt
new print with the #29 AWG>
Attachments
rewind- cut all the insl first so you do not have to stop after every section,
cut a piece of cardboard that fits snug into the bobbin, use that for your measuring stick, draw some guide lines and use a new blade to slice 3 mil Nomex layer insl.
using Nomex instead of KP means 3 KV strength instead of maybe 1 KV depending on the paper.
some people say the Nomex instead of KP actually changes the sound for the better.
increase in coil build as you wind will crate need to cut longer and longer strips off the main strip, use MLT spec as a guide line and leave extra, you can trim excess if needed when you tape it to the coil,
used the build chart to figure out how many pieces you need,
cut a piece of cardboard that fits snug into the bobbin, use that for your measuring stick, draw some guide lines and use a new blade to slice 3 mil Nomex layer insl.
using Nomex instead of KP means 3 KV strength instead of maybe 1 KV depending on the paper.
some people say the Nomex instead of KP actually changes the sound for the better.
increase in coil build as you wind will crate need to cut longer and longer strips off the main strip, use MLT spec as a guide line and leave extra, you can trim excess if needed when you tape it to the coil,
used the build chart to figure out how many pieces you need,
