easy transmission-line transformer

formatting link

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin
Loading thread data ...

Pot_Core_TXline.JPG

For those of us who are not mind-readers, what frequency is this designed for? I can't make out what type of connectors those are, either.

Reply to
Cursitor Doom

Pulses, maybe microseconds wide, risetimes around 1 ns maybe.

The connectors are UMC, sometimes called UMCC. The cables can be bought in assorted lengths from Digikey.

formatting link

formatting link

This will be really easy to assemble. No stripping, soldering, any of that manual labor. A screw through the PCB will secure the pot core. Maybe I don't need the bobbin.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

line.JPG

I did that back in 1978 and pubished it in 1979.

formatting link

I didn't think it was worth publishing but Ken Ghiggino wanted the publicat ion. His first draft wasn't all that good, but it polished up nicely. Dave Phillips was supervising Ken at the time. By the time I'd finished the pape r I was sure that I should have used an emitter-coupled monostable, but wha t we had put together worked fine in our application.

--
Bill Sloman, Sydney
Reply to
bill.sloman

At small amplitudes and modest impedances (~50 ohms), it'll go down to the ~100kHz range. Maybe lower, especially if the source impedance is lower. The limit will be higher at large amplitude, due to saturation.

Upper bandwidth is limited by electrical length, for this type of TLT. The first resonance is 1/2 wave. It's not obvious how much TL is wound onto the bobbin, but I'd suppose 1ns is quite reasonable (which is what JL's looking to do).

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

This will be for fast pulses, numbers like 50 volts into 50 ohms for a couple hundred ns. I've got to build the entire driver and see what happens. This pot core, one that we have in stock, is ungapped, so has lots of Al but will saturate easily. Adding the transmission-line length factor makes the magnetics design yet more complex. I'll just breadboard it and set my specs at whatever looks nice.

Digikey sells the connectorized cables in various lengths, in about 1" increments, so it's easy to change. Manufacturing doesn't like the way we're doing it now:

formatting link

formatting link

I suppose they do have a point.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

Are those the SSMB connectors you said you like?

Reply to
Clifford Heath

Better amplitude with a powdered iron core? What low frequency cut-off might be expected then?

Clifford Heath.

Reply to
Clifford Heath

Those are SMBs. They snap in, mate and unmate instantly. They are good for maybe 10 GHz. They are cheap, $1.75 roughly.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

ned

he

.

The

the

ing

In as far as the transmission line has equal currents travelling in opposit e directions in the inner and outer, it takes a while before the much flux starts to build up in the core.

You'd need a long pulse to get the core anywhere near saturation.

Coax connectors aren't cheap. Solder connections are. The virtue of the coa x connector is that it is a great deal more symmetrical, which is to say in troduces a whole lot less extra inductance.

We ran into this at Cambridge Instruments in a completely different context .

The 30kV supply to the electron gun was routed up to the gun on high voltag e-rated coax. As long as it was terminated at the gun with solder tails, a "flashover" at the gun produced hundred amp currents in the ground connecti ons, and voltage surges that forced us to fit protection diodes and resisto rs between boards.

When we finally went over to a proper coaxial high voltage connector at the gun, the flash-overs didn't produce any significant return currents in the ground connections. The high voltage coax connector wasn't cheap, but we'd spent a bundle dealing with the consequences of fashovers, and as much aga in adding surge protection to the inter-board connections.

If we'd gone for neatness earlier on, we'd have saved quite a bit.

--
Bill Sloman, Sydney
Reply to
bill.sloman

Higher.

Powdered iron has much lower permeability. It's suitable for inductors (that store energy, regardless of the number of windings -- call those "coupled inductors"), not transformers (which aren't intended to store energy, or equivalently, are supposed to have a high parallel impedance at most frequencies).

What you need is more cross sectional area, higher Bsat, and to some extent, high permeability. Ferrite is pretty much all the same (Bsat in the same

0.3 to 0.45T ballpark) -- what you're buying is different permeability * frequency (it's a tradeoff, like GBW). Nanocrystalline is the best, with more Bsat (0.8-1.5T), and awfully high permeability (over 20k). It's a magical material, with permeability surpassed only by permalloys, but only at low frequency (which have been used for audio microphone and isolation transformers).

Permeability only needs to be high enough to satisfy the parallel impedance (magnetization current) spec; it's usually burdensome below a few hundred, with >2000 being very typical for power transformers (everything from laminated iron to ferrite following this coincidence).

Among powder materials, the highest permeability grade MPP is marginal for transformer use -- but I doubt it gives any more bandwidth.

Tim

--
Seven Transistor Labs, LLC 
Electrical Engineering Consultation and Contract Design 
 Click to see the full signature
Reply to
Tim Williams

I like MCX for almost the same reasons. Tried different types once and for some reason (which I don't remember now) I preferred MCX...

Pere

Reply to
o pere o

OK those look more like what I expected. So mostly this is about connectorized cables.

George H.

Reply to
George Herold

MCXs are good too. The SMBs are kinda standard in aerospace instrumentation, so we mostly use them; they do detent pretty well, a bit snappier than MCX. SMAs are better for really fast stuff, but are a pain to mate amd unmate, and have to be spaced farther apart than SMB or MCX.

--
John Larkin         Highland Technology, Inc 

lunatic fringe electronics
 Click to see the full signature
Reply to
John Larkin

ned

he

.

The

the

ing

looks like the ID of those cores are big enough to fit cables with connecto rs

Reply to
Lasse Langwadt Christensen

Is that paper 1/4" grid with a faint 5mm grid?

Never seen the like in all my days, though I did once see a surveyor's tape with feet and suspiciously large inches, which turned out to be decifeet.

Cheers

--
Clive
Reply to
Clive Arthur

That's an astronomer's tape, graduated in attoparsecs. (1 aPc = 1 decifoot to within astronomical tolerances.)

Cheers

Phil Hobbs

Reply to
pcdhobbs

It's a 4x4 grid (0.25") on one side and 5x5 (0.2") on the other. Nice to scribble on.

--
John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  
 Click to see the full signature
Reply to
John Larkin

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.