Home made feed through terminator for BNC

[david77]

While in the final stages of fixing an old frequency generator it turned out I needed a feed through terminator to terminate the output of the generator into 50R while adusting the unit.
Of course I did not have one and it was two in the morning and I wanted to finish this thing.

What do you do? First I took a BNC T plugged one of those BNC to banana adapters in one end and hooked two 100R resistors to the terminal posts. Not ideal but worked well for frequencies under about 1MHz.
Now that generator goes up all the way to 11MHz and there the rectangles coming out didn't look like rectangles at all. In fact I had a hard time triggering the scope on that. Another solution was clearly needed.

I finally came up with a neat solution. It turns out that ordinary panel mount BNC sockets fit into ordinary solder type BNC plugs, the thread is the same. At least in the types I had in my drawer.

I soldered three 0805 150R resistors between the centre pin and the ground to get to the desired 50R impedance.
The pin of the plug is soldered straight onto the solder cup of the socket with a wire to give some stability.



The finished product looks like this:



Probably not ideal but it was a cheap solution that helped me finish the adjustments on my function generator.
It's now in my drawer waiting for future use.

[Galaxyrise]

Oh, that's a nice DIY! How's it compare to the real thing?  How fiddly was it getting those resistors on there? Did you attach them before or after the pin?

[david77]

Can't really say, I don't have the real thing.
I soldered the pin first the a piece of wire sticking up from the ground connection. Then I placed all three resistors in between there with tweezers and tacked them on to the pin.

[c4757p]

My suspicion is that it works just as well at least to several hundred megahertz.

[branadic]

You have better used a star arragement for the three resistors to minimize stray capacitance between them. But of cource, this is the old fashion style feed trough solution.

My suspicion is that it works just as well at least to several hundred megahertz.

What needs to be proved.
The HZ22 for examples was primarily designed to work for scopes up to 100MHz:

http://www.mikrocontroller.net/attachment/98198/HZ22.jpg

They now sale it with the info it is usable for up to 1GHz. If you put it to a VNA or measure it with TDR you would be shocked about it's behaviour.

[Galaxyrise]

But of cource, this is the old fashion style feed trough solution.

What's the new fashion?

[robrenz]

Very nicely done!

[branadic]

What's the new fashion?

To compensate the input capacitance of the scope? The disadvantage is, that each feed through termination is for a specific scope, the advantage is the higher bandwidth compared to the shown principle.

Or more simple, use a scope with internal termination.

[david77]

Well, I'm not claiming anything so I have nothing to prove. I just thought it's a neat solution and wanted to share it.
If someone here has the means to put it through some tests to see how this quick and dirty terminator behaves I'd be very intersted. I do not have the neccessary equipment to characterize the thing.

[Tuomas]

Nice find, I happened to have similar parts lying around and have a need for an inline terminator as well.

My panel mount connector was slightly different, the insulator extended longer than what you had in your picture, so I couldn't directly solder the resistors between the center pin and shield. I cut a small circular piece of 0.3mm thick copper, drilled a hole in the middle and soldered it as a "hat" to the center pin. Then I soldered four 0805 200ohm resistors between the "hat" and the shield. The resistors stayed very well in place between the two parts so soldering was quite easy. Then I wrapped the soldered area in kapton tape and shoved it inside the other connector.

I would imagine having that extra copper has incerased capacitance and I'm sure there are other issues as well, but it seems to work. I only need it for frequencies <100MHz so I'm assuming it should probably do a passable job.

[G7PSK]

I wanted a bnc terminator a while back, all I had was some bnc crimp on plugs. I made a half watt 50 ohm with a couple of 100 ohm through hole resistors and covered in shrink sleeve and another one using smd resistors the wattage of which I do not know but cant be much as they are so small. 

[grumpydoc]

Resurrecting this thread a bit.

Inspired by the original post and the fact that a) I needed another terminator for some 'scope calibration b) the shops were shut  c) I'd just found a reel of 0.25W 200ohm 1206 resistors in the spare parts bin and d) the cheapest feedthrough terminator I could find on fleabay was still twenty quid, I thought I'd have a go at building a feedthrough terminator from a spare panel mount BNC socket and clamp-on plug myself.

To add to the gallery here are some photos of the result



And the top view


The finished article looks just the same as david77's

I'd like to be able to measure the performance of this properly but don't have the kit - however I can say that it doesn't appear to be especially stellar. Probably OK to a couple of hundred MHz but at 500MHz both this and a cheapo 50ohm resistor soldered to a BNC crimp plug (plus tee) both showed about 6dB loss compared with my scope's internal 50ohm termination.  Worse, with either connected to the 'scope, my signal generator actually complained saying "RF amp failure or unterminated" when I tried to drive them at 1V RMS - presumably the VSWR is bad enough that the sig gen sees it and sulks - from my rough understanding of VSWR I'm assuming that if the 'scope only sees half the voltage the sig gen is putting out then the rest is getting reflected so the VSWR would be 3:1 

I confess I did expect that it would be better than a tee plus terminator given that they don't need the tee - perhaps my construction is rubbish?

So, OK in a pinch and as long as the frequency isn't too high but not a substitute for equipment with correct internal termination. I might make a couple more for low frequency use as they're more convenient than having to use a tee

[PA4TIM]

The best configuration is 2 x 100 Ohm soldered opposit from each other. This minimizes paracitic capacitance and inductance.
 
Nut for the rest nice done.

I have done some tests on feedtrough terminators with a vna and TDR. The best result was a attenuator in series with a in line terminator. This because the input impedance of the scope is decreasing with frequency. (how much is depending the brand and model (my former Rigol was only 30 Ohm at 100 MHz) Best is a scope with 50 Ohm input.

[grumpydoc]

The best configuration is 2 x 100 Ohm soldered opposit from each other. This minimizes paracitic capacitance and inductance.

Yes, that's why I spaced the resistors out. I didn't have any 100ohm in the parts bin so was stuck with the 200ohm ones. Well, some 1/3W through hole but they were much too large physically.

[babysitter]

Inspired by your posts, I fabricated a limiter out of:
* 2 x unknown SOT23ish RF NPN transistors
* 1 x N plug
* 1 x BNC front panel mount BNC socket

Exchanging a N-BNC Adapter in a small bridge from TG to SA for transmission measurement gives a flat trace up to 150-200 MHz where it started to bobble up and down, at ~500 MHz the bobbling becomes > 1 dB.
More exact measurement when baby lets me !

[BravoV]

Bumping out an old thread.

Ignore the BNC connectors, just put there as size comparison and I will find more suitable ones if needed.

As you can see by cutting off both flanges left & right, the resistor will fit in nicely when using housing like above examples.

My question is whether this particular termination resistor is "better" than ordinary resistor ? The datasheet of the resistor attached.

[T3sl4co1l]

No, because you can't heatsink it (without an awful lot more metalwork than just cutting the tabs!).

Tim

[rs20]

+1, and also it wouldn't be a feed through terminator, as the resistor only has a contact on one side (unless you bridged a wire over to the other side, by which stage your RF performance would be vastly worse than standard surface-mount resistors in a radial pattern.)

[BravoV]

No, because you can't heatsink it (without an awful lot more metalwork than just cutting the tabs!).

Tim

+1, and also it wouldn't be a feed through terminator, as the resistor only has a contact on one side (unless you bridged a wire over to the other side, by which stage your RF performance would be vastly worse than standard surface-mount resistors in a radial pattern.)

Ok, noted, thanks. Btw, that RF resistor I got for almost free though, I guess it will be stuck forever in the drawer.

[T3sl4co1l]

+1, and also it wouldn't be a feed through terminator, as the resistor only has a contact on one side (unless you bridged a wire over to the other side, by which stage your RF performance would be vastly worse than standard surface-mount resistors in a radial pattern.)

Well, if the impedance of that link is significantly different from 50 ohms, absolutely -- but that's a matter of geometry, not a guarantee.  So I don't see that being automatically bad.  Just don't use tinsel wire.

Also, but...so... it's nearly a useless matter anyway, because the whole point of a feed-through terminator is to have a stub length on the far end of it!  There is the length of the 'output' side connector at least, plus the length of the connector it plugs onto, plus whatever else is beyond that (i.e., the oscilloscope's 20pF equivalent input impedance, internal cable and trace lengths..).

So the best you can hope for, is to minimize that stub length, and to keep its impedance near nominal (say, 50 ohms give or take a not-too-gross margin like +/- 6dB).

The best possible design, would be to have the termination resistance as close to the 'output' side as possible (minimizing stub length), and to have the resistance as resistive as possible, over as wide a bandwidth as possible (ideally, a coaxial, disc shaped bulk resistor?).  Probably with an R+L, R||C snubber to dampen the following stub length (assuming its length is known).  Which I suppose is what the best designed units might do.  But even if you're using a short leaded resistor instead of a perfect coaxial one, you're not dramatically increasing the stub length(s) -- it's only a cm here or there, when the overall stub length (including external parts) might up to 10cm or so.

And if you really are working with frequencies so high that 10cm of stub length is critical (i.e., about <1ns or >1GHz), you have no choice but to go out and get internally terminated equipment!

Tim

[vindoline]

I'm interested in making up a set of these DIY terminators. However, all the BNC jacks and plugs in my parts box have different thread sizes  . The jacks are 3/8" and the plugs are 7/16. Can anyone point me towards comparable parts?
Thanks a lot!

[bjcuizon]

Just here to resurrect this thread a bit and maybe inspire others. 

I also built this DIY feedthru terminator. Firstly, because I needed to get clean signals from my fun gen and secondly it was a cheap solution.
I put two Bourns 100ohm 1W (2512) resistors in parallel to minimize stray capacitance and I also 3D printed some plastic sleeves that go around the BNC connector and labelled it so that it looks like a bought one. 
Here are some pictures of what I did...
Overall, this was a successful project.

[BravoV]

Nice ! 

I ended with buying cheapo but full metal BNC 50 Ohm straight thru terminator at Aliexpress  , physically not bad imo, posted -> HERE

[rs20]

Very nicely done! I can tell from your pictures, though, how you made a electrical + strong mechanical connection between the two outer ("shield") conductors?

[bjcuizon]

When I printed the plastic part, it did not fit over the male bnc just yet. So I had to sand down the inside just to make it fit tight enough. The female bnc locked tight enough too because I effectively eliminated the gap between the two connectors (as what the others have above in their pictures). So when I disconnect a bnc plug from the female bnc, it does not turn with it.

On the electrical side, the female bnc just screws into the male bnc. In that way, it just connects each other quite nicely.