new "class x" transformers from Edcor

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

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

edanderson

Well-known member
Joined
Jun 4, 2004
Messages
637
Location
atlanta
i haven't tried these yet, but i've been very pleased with the other units i've gotten from edcor, especially for the price. now they have a new line of tube amp output and power transformers, plus the XSM series of line matching transformers. the XSM series looks to be built on 625 size lams, so it would be the same size as the a*p*i output, and most of the units by jensen, cinemag, etc.

http://www.edcorusa.com/classx/

they've also thrown in some info about the m6 based ribbon mic transformer they now offer. someone around here used one.

edcor makes some great transfomers for the money. for experimenting, i personally don't think you can go wrong with ordering a few common types to have on hand.

ed (no relation)
 
Rock on, Edcor!

Their main business has been installed sound, but they must have noticed the uptick in orders from DIY audio types, and responded accordingly. This is very cool.
 
Man, a 25W P-P output transformer for 18 bucks... Edcor kicks so much ass!!! :thumb:

Peace,
Al.
 
They have a 10K:600 interstage transformer that, according to the spec, can handle DC in the primary and would possibly make a nice plate-coupled output for a tube preamp (e.g., something like the RCA BC-2B pre).

I like how their stuff is prominently labelled "Made in the USA!" (complete with exclamation point).
 
Of all the cheapass, DIY misers I have ever met, ....you guys......take the cake!
:razz:

Shipping will be more than the profit they make on that transformer. Some slave winder is really paying their dues. I mean, to be a winder, you have to be on drugs. You have to have a radio at your station. It is soooooo boring!
How can you afford good drugs at todays prices when you are charging 18 bucks? Must be a relegious company. Holy Ghost Power! Or people from down south, like the shop right down the street from me. Don't expect to see OSHA approved ventillation systems near the varnish tank and dry out oven at the Edcor factory. Someone is busy getting lung cancer so you can have your 18 dollar x-former, so enjoy it for all it is worth.
Yes, I worked in a transformer shop for ten years, so I know what I am talking about as far as winders are concerned.

I would hate to work in a restaurant and rely on your tips to pay the rent.

Seriously though, I just swept a cheapass 5:1, 600 to 120 volt small power transformer for a product at work, and the dang thing was flatter than a pancake. This is just pri-sec construction on a cheap core, so go figure. Maybe there are other things to consider besides response, like excitation current, etc.

cj
 
Yes. The 2503 uses multistrand wire consisting of 4 conductors that are twisted together to provide a leakage coupled perfect phase rolloff at about 6 gigahertz. The edcor will be shifted during its' 80 kc nosedive.

cj
 
Those specs really don't mean much, since they don't specify conditions. But for what it's worth, the WSM series has measured very well on my bench under typical conditions of source and load impedance.

I suspect that if the HF corner of the "X" series really is lower, it's because they had to use more turns for a given primary inductance (because of the air gap). And more turns equals greater distributed capacitance.

Besides, can YOU hear when your signal is down 1dB at 20kHz? :razz:
 
I just ordered four samples from the XSM series. I'll let ya know what I think when I get them.

I want to try the XSM10K/600 in a transformer-coupled version of my "one-bottle" preamp, replacing the plate resistor and coupling cap.
 
So specs and prices aside, does anyone have any comments on how the WSM series sound? I just ordered a pair of 600:600's for my new PM1000 modules and also plan on ordering a pair of high end units, like Lundahl or Cinemag for comparison. Should I expect a big difference, noticleble difference, or negligible difference? BTW, I can't seem to find any specs for the lam material.

Ultimately my shoot-out will answer my questions, but any comments will also be appreciated.

-Chris
 
NYDAVE!!

here's a quick one for you... this is probably more drawing board material. If the thread gets out of hand, I'll start a new one over there....

When you replace the plate resistor and coupling cap in your design and send the B+ through the primary of the output tranny, does the DC resistance of the transformer become the plate resistor more or less? I've never done an output like this... only with a buffer stage AC coupled to a transformer. Mostly because it's a little bit cheaper. big-ass output iron that specs DC through the primary always seemed on the expensive side to me. Until edcor offered this... If it works out, I may try an amp like this.


joe
 
As a first approximation, ignoring hopefully "minor" parasitics: at AC, the plate sees the primary inductance and the reflected (stepped-up) impedance from the secondary in parallel--and this is in series with the copper resistance. At DC, it's just the copper resistance. Since the small copper resistance drops a lot less voltage than a plate resistor, the transformer is not a drop-in replacement. You have to redesign the output stage for the new DC conditions.
 
chris:

all edcor transformers are made using grain oriented M6 grade silicon steel. that is the only core material they use.

it really depends on what you order from cinemag as to how much of a difference you would hear vs. edcor. the WSM series is made on a smaller size core than the cinemags, and at cinemag they like to use nickel in their cores. you will probably find it easier to push the edcors and get "transformer" sound out of them. the bass will sound different as the level increases.

all the lundahls will sound different from the cinemags and edcors. they are constructed in a different way. if you are looking for a real contrast to the edcors, i would say get a pair of lundahls.

ed
 
i see. can you make up the difference with a resistor? is it common for the resistor to be on the b+ side or the plate side of the primary? or would you re-design completely? A lower plate resistance value would make a lower output-Z at the expense of voltage gain for the stage, correct? just trying to make sure I'm on the right track here. You were using voltage feedback in your 'one-bottle' design, i think. you would still tap into the same place (between the plate and the tranny, instead of the plate and the plate load resistor) for the feedback? Occasionally I see feedback coming from odd places on output transformers, it may be from a power output stages I'm thinking of, though.



I'll copy and paste some of this into the drawing board forum, and we can continue this discussion there. You guys can have your edcor thread back, now. I'm planning on ordering a couple of these, btw. hopefully, they'll work out.

joe
 
I received my samples today and I made some preliminary measurements. Note that the inductance measurements are with NO DC applied.

Model......................Pri. Inductance (H)... Pri. DCR..... Sec. DCR
XSM600/150....................16.......................72.................38
XSM600/600................... 17..................... 117..............118
XSM10K/600................... 28..................... 140................36
XSM10K/10K....................31..................... 225..............223

The high DCR of the 600/600 model is disturbing. If you actually terminated the secondary in 600 ohms, you'd lose 3dB just in the transformer! Ya might wanna talk to them about that if you plan to order one to use in a true 600-ohm system. They're pretty accomodating about special requests.
 
> The high DCR of the 600/600 model is disturbing.

ALL the numbers look wrong.

You expect DCR and Inductance to scale nearly as rated impedance: not even close.

Total DCR in a good design will be 5% to 20% of rated impedance; each winding should be half or 3% to 10%.
72/600 is 12%, 38/150 is 25%
117/600 is 20%
140/10K is 1.4%, 36/600 is 6%
225/10K is 2%

Are the voltage ratios right? If so, the "600/150" may have both windings the same gauge, using 66% and 33% of the area, instead of 50%-50% optimum.

ASSuming 20Hz, the inducances are all over the place.

16H @ 20Hz is 2K, "too good" for 600 ohms (though 600Ω trannies are often wound this high).
30H @ 20Hz is 3.8K, not good enough for "10K" at 20Hz; this should be rated 53Hz.

Makes you want to peel one apart and recompute the windings, TELL them what to put on.
 
I measured the primary inductance of the WSM10K/600 using the method described in the "Measuring Inductance" thread on the drawing board. I used 20Hz and 50Hz at a level of +12dBU and varied the direct current through the primary in steps from 0 to 40mA. I'm afraid the results are not encouraging for applications that involve more than a couple of mA of unbalanced DC through the primary. See chart below.

Chart

Interestingly, my bridge method gave a much higher figure for the "no DC" primary inductance than what my handheld LCR meter had been telling me. Sadly, this nice high inductance absolutely plummets as you draw direct current through the primary. This transformer should be OK for a reasonably well-balanced push-pull output, though.

Edit: Don't take these results as gospel yet because I'm starting to have some misgivings about the test method. Just accept them as "preliminary" data for now, OK? :wink:
 
[quote author="NewYorkDave"]Edit: Don't take these results as gospel yet because I'm starting to have some misgivings about the test method. Just accept them as "preliminary" data for now, OK? :wink:[/quote]

It is much appreciated though. :grin:

Thank You,
Tamas
 
Your test level, +12dBU on the 10K winding, is just 0dBU or 0.775 Volts on the secondary, 3.16V on the primary.

Assuming nominal impedances, 3.16V across 10K is 0.316mA RMS, 0.45mA Peak.

To make 0.45mA peak swing, we have to bias the active device to 0.45mA minimum, at least 0.5mA and probably 1mA.

At 1mA DC, the inductance seems to be 47H at 20Hz. 47*20*6.28= 5,900 ohms, not 10K. At 50Hz, 32H, we do just touch 10,048 ohms.

So between a Pentode and a 600 ohm load, we may get 50Hz at -3dB.

Between a 10K Triode and a 600 ohm load, we may get a bit better than 20Hz.

At 1mA bias, the peak output current (if the core can stand more than 0.45mA peak AC current) in 600 ohms is 4mA or 2.4V peak or 1.7V RMS or +6.8dBm.

I dunno. I think I got mike transformers that work better than these.

I think they want many more turns on the primary (Edcor may not stock a wire thin enough), and perhaps a bit of a gap (maybe just odd stacking).

Perhaps primary gauge could be 0.7 the diameter, 2X the turns. Secondary gauge should be twice that diameter, 1/4 as many turns as the primary. The sum should fill the window over 80% or so.

That would seem to give a no-DC primary resistance of ~560Ω, inductance over 100H. I fear however the inductance will fall at a uselessly low DC current.

This core is 2 inches wide?
 

Latest posts

Back
Top