Using Lundahl 7903 as Output TX... -pinout???

[SSLtech]

I'm not finding the Lundahl data sheet for the 7903 easy reading in terms of how I need to configure the windings in order to test one as an output TX for an LA-2a...

Can anyone tell me which combination of windings I'd need to make it happen?

I believe that the 7903 is a 1+1+1+1 : 2+2+2+2 voltage ratio.

The datasheet is here: http://www.lundahl.se/pdfs/datash/7903.pdf

Which combination of L1 L2 L4 and R1 R2 and R4 do I need?

Thanks everyone! :thumb:

Keef

 

[CJ]

What, your having problems with this simple hookah? I mean hookup?
No wait, that's Steve, who's probably puttin on his spurs right about now, boy howdy. Where's my Skoal bandit? Still watchin Gunsmoke re-runs, Steve? Ooohh, that Miss Kitty! I always wanted to put the pork to her! :grin:

You got me on the hook-up. But the idea? Genius! I was wondering what I was going to do with that paperweight, as it is not my favorite input transfiormer, that's for sure.

Here's what it looks like on the inside, just for the heck of it. WeCo style lams, nice puter made windings, dipped, whipped, and ready to rip:

Let us know how it sounds, and don't drop it, the pins are brittle, and thats a lot of Ni for an input. It would break your big toe and then how would you heel and toe that Volkesensputtin around Whore-lando?

Man they ran the Grand Prix down here in San Jose yesterday. You could here them practicing at 4 in the morning. But what a joke. The course was too thin to pass, so the order never changed. And they ran it over the light rail tracks so the cars were getting air, comming down hard, and not finishing. But the beauty pagent was alright!

And again, I great idea! (why didn't I think of that? :? )

 

[Scenaria]

L4 R4 will give ya +28dbu this would be my vote

if you want to double check... PM kevin over at K and K (thats where I got the lundahls)

http://www.groupdiy.com/forum/privmsg.php?mode=post&u=40

 

[SSLtech]

but L4 R4 will give you a 2:1 ratio, no???

I want more..... -MORE!!!!!

Keef

 

[CJ]

See if you can get away with floating 13 and 16.

 

[kevinc]

I suspect that you want 4:1, huh? Use R4 (pri) - L2 (sec). That gives you close to what an A24 would in an objective sense, but really kicks butt in a subjective sense!

 

[CJ]

Use the sec as your pri, yes, as it is the inner windings and thus more voltage isolation from hi ac signal to ground/core/case and also less leakage to case.

Each sec section of the 7903 is about 52 Henries, so R4 is going to give you about 800 Henries, where as an A-24 probably has about 250 Henries in 4:1. No biggy, you might have more bass, but will need a bigger output cap to deliver said bass and match up with the inductance a little better. I do not know what the resonant freq will be with the stock cap and 800 henries.

R2 would give you about 208 Henries, closer to the stock A24. You would also have better humbucking, but this is not really important in a non push-pull output stage.

The pri of the 7903 has extremely low dcr which means huge wire ( I can see it from here, folks!) for a mic input, which is what Per probably means by "extradorinary requirements". Like lightening hitting the primary, Per? So in theory, you would want this big wire on the bussiness end of the output stage, but again, no big deal, as the sec wire is pretty hefty also.

I get a "D" reading of about 1.003 on the primary, which is about the highest I have seen among the stuff I have fooled with.

Play around with different hookups, thats the best way to outsmart us geeks.

You can snip pins 19 and 22, since they are NC's. This little "mod" makes life suprisingly easier.

I just got a whopping 541 pf between pin 16 (shield) and coil 2 sec, so....
this is the tradeoff you make when you use monster cables for magnet wire. But maybe the tricky hookup configs help neutralize this.

 

[CJ]

sweep test, L4-R4 and R4-L2, both about the same, major bump starting at 1 khz and shelfing at around 10k, dips way down a lil past 100k, then whamo, big peak up to 1 meg.
hmm, maybe this one is designed to be rugged, but not a 1538xl,...... but give it a sweep and see what it does on your end.

 

[Gus]

FWIW I used one as the transformer in the 800 type microphone I built as 8:1 out.

Like CJ posted don't push hard on the pin side and if you mount it to a PCB or perf board I would use tie wraps to hold it to the board because of is mass.

 

[kevinc]

Chris,

What were your source and load impedances?

 

[SSLtech]

I'm definately having some problems here...

The LA-2a is working perfectly well all the way up to the negative (outgoing) plate on the cap feeding the output TX.

If I disconnect the cap from the primary (what would have been the secondary) of the7903, I get excellent, strong signal. As soon as I connect the cap to the TX, I get big signal load-down. I've got a UTC A-24 sitting round here, if I don;t get anywhere soon I might just wire it in to prove that the circuit is okay (it scopes out the same clipping limit as another LA-2a I have here... removing the cap from the primary makes no significant change to the signal level there... outputs unterminated in both cases.)

It's almost as though there's a shorted turn in there. With no load or connection across the secondary (would have been the primary) then the signal hits the deck as soon as you load the cap output with the primary. It loses over 20dB in absoulte level and distorts horrifically.

-There wouldn't be a mistake on the data sheet wiring would there? I'm using L2 R4 with R4 as the primary and L2 as the secondary.

Keith

 

[SSLtech]

Okay, I figured out the problem!

-I did indeed have an effective short across part of the windings. Pin 24 was shorting to ground through the hole in the metal plate to which it is mounted. Since the hole is sealed with hot-glue, I had no option but to re-assign pin 24 as the low pin (which would therefore be connected to ground anyway) and re-sequence the windings to maintain the "tricky hookup configs" that CJ makes mention of. -That took some mental gymnastics! I also tried a UTC, which will pass a much higher level. Neither is loading down the primary side, even at low frequencies, but the UTC is better at high levels... which is not what I expected at all!

Anyhow, I'll put it up on the Neutrik tomorrow, then later on I'll take it to Scenaria's for a spin on the AP. He's got a couple of others to compare it to, including Sowter and UTC. At ome all I have is a UTC to compare it to.

We'll post some thoughts when we've got some data to back it up with.

Keith