If you run a cable AC sim 1 Hz to 1GHz and step the load resistance from 1k to 1MEG its as flat as a pancake (current and voltage) up to about 1 MHz. After that you get the typical resonances at high load resistances up to c. 10 MHz and then a steep drop off - I used ideal components, but, but otherwise nothing else.
I am still failing to see how a 10 kHz sine wave can be reflected down a down a 4 metre resistively terminated cable such that the reflected wave is so severely phase shifted that its c. 180 degrees out of phase with the source signal (CB's scope plot).
Anyway, I'll step out of this specific discussion because people are either having fun (I hope) or they are secretly working for the cable industry.
I am still failing to see how a 10 kHz sine wave can be reflected down a down a 4 metre resistively terminated cable such that the reflected wave is so severely phase shifted that its c. 180 degrees out of phase with the source signal (CB's scope plot).
Anyway, I'll step out of this specific discussion because people are either having fun (I hope) or they are secretly working for the cable industry.
CB had 53dB directivity at 1Khz, 30 dB isolation.
But you're right, I didn't "see" him personally do it..😉
Is there a reason for you to attempt to misquote? Seeing a reflection in a cable doesn't mean the cable did it, just that you actually have to look at the signal in the cable. I mean, you could try to look for it in the freezer with the system pictures, but I suspect you won't find them there..
I suspect you are talking to a line with an open termination..😀
Yup. Also, I found back in the day that there were significant initial condition errors as well. Not to mention the issue of mutual coupling.
And ignored in all this nonsense is the fact that I specifically stated up front that those who do not like or understand t-lines at sub wavelength can easily use lumped elements, as they produce the exact same results.
jn
Um, negative reflection coefficient ...
Re-examine Scotts example of an extremely long line, and the last 4.9 meters.
How does an extremely long line know to not reflect when we look only at the last 5 meters?
It is fun watching the deflection and posturing when someone doesn't understand the arguments.
It's also fun to see you accuse me of possibly conspiring to gain financially via arguments, it diverts from the possibility that you could be wrong.. and that I'm a bad guy..
jn
Seriously though, engineering approximations (no matter how good)
do not negate the actual physics involved. Ask Newton about that.
At this rate, engineers will just simulate, instead of actually building stuff.
Oh, wait.
Last edited:
Nice. CB used a pair of 10,000 u ferrite toroid cores, 36mm ID, 22 mm ID, 15mm long, with 31 turns. Winding inductance of 22 mH. Primary was single turn.
I suspect the actual schematic is generic..on one page he calls it a two transformer directional coupler and "often used for RF measurements" because it absorbs no power. In the schematic he calls it a dual directional coupler. Also, it just requires two resistors to match the cable impedance.
OH, almost forgot..CB's coupler was a 4 port device, your stated device is a three.
jn
Last edited:
I only think you are bad because of your method 😎 Still whatever turns you on, did someone call this your hobbyhorse? 😉
I wonder at what line to wavelength ratio a mismatched end termination stops reflecting..
jn
Yah, I gigged him for that.😉
And so funny that this diversion (how t-lines work) is away from the crux of the discussion. But then again, this is the blowtorch thread..
jn
Last edited:
Hmm, I'll have to reflect on that.
https://images.immediate.co.uk/prod...y=90&crop=7px,73px,926px,399px&resize=940,404
Oh man, I stepped right into that, didn't I😱
ah, found the coupler...
The tandem match - a closer look
with that schematic and the transformer information I just provided, we have exactly what CB did..
Now the smart guys here can figure out if he was nuts, or waay ahead of his time..aaaand, we can exactly duplicate what he did. Bingo.
jn
Last edited:
When I made my measurements I used a resistor load as I was testing cable constructions. I had LRC substitution boxes making up my load balance. A few clicks would quickly get a decent balance. I did make my own LRC boxes in a style to look like they were quite old. Binding posts that also took banana plugs etc. Probably now have had them 20 years.
JN - just a thought - set up a sim and show what you are saying - sometimes a picture really is better than words..?
No posturing and no deflection. I did a sim with just a cable as I described.
If it’s the load causing this why talk about ‘the cable’? Why not separate load from cable as I suggested and then do the analysis? Surely then the conclusion would be below a MHz, the cable plays little or no role and the load reactance at the other end of the cable dominates.
We could perhaps then have a sensible discussion about how to mitigate crossover reactance (Zobel?) rather than obfuscate the issue and feed the cable fantasists. Or dredge up CB’s test results with what was clearly a flawed amplifier.
Anyway, I already came back into this discussion when I said I would stay out. Better to leave it and draw our separate conclusions.
In my gallery, there is a graph showing settling times for a 100 ohm zip cable 20 feet long, feeding loads from 2 to 20 ohms.
I do not know how to link to my gallery from an IPad.
For a specific length cable, when it is loaded well below it's Z, it will take time to settle to final value.
This can be modeled by t-line understandings, or it can be modeled using a distributed LC model. Both results are identical.
Note that for the variation of load, the settling time varies drastically. At the worst, 2 ohms, it takes about 10 uSec to reach 80%.
This is not a long cable, and it is lower z than typical zip, so these numbers are lower than reality.
This is the test that needs to be run.
Jn
Last edited:
I am quite surprised there is no thread on cable content. Not sure if it's me or them.. I find How it's made, Aerial disasters, Mega machines, omg even original Star Trek series episodes interest me more than the bazillion channels I get..
Jn
- Status
- Not open for further replies.