|
Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets. |
|
Thread Tools |
18th Nov 2017, 1:16 am | #1 |
Octode
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,795
|
Minature Coax cable
Hi Peeps. I bought some "unmarked" minature coax leads with SMB/SMC plugs attached. At the time I thought these were excellent value at £1 each and approx 4 ft long. I assume the cables are "ex cell phone cabinet equipment".
Recently I have come to doubt that they are 50R impedance and wonder if any of you learned members have used these and what conclusions you made. The reason I bring this up is..I am trying to make a "flat" response amp 3.5 Mhz - 30 Mhz for the Level meter for my SSB RF generator...and I find that I keep getting conflicting results of certain frequencies being either + or - say 3 db depending on how the amp is connected. My Rogol Spectrum analyser says the cable is "flat" but when connected to the Marconi 2022, "terminated with 50R" it is not. I am using a MAR-6 as the buffer amplifier and a 2SC388 on the output of the MAR 6 as an emitter follower. The Rogol also says the MAR 6 is flat, but the reading of the meter is not consistent. Using 50R RG 285 cable, gives much flatter results, but is much more bulky. I just wonder if I am making a big mistake using these cables.
__________________
Should get out more. Regards Wendy G8BZY |
18th Nov 2017, 3:32 am | #2 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Minature Coax cable
Hi Wendy,
Just some ideas: It might be that there is not enough buffering between the output of the MAR-6 and the rectifiers. One way to be sure would be to replace the existing emitter follower with an AD8056 OP amp wired as a follower with a suitable resistors to set its input impedance to suit the MAR-6. Possibly also a small resistor, such as 4.7R in series with the emitter follower output (or OP amp if you use one) to limit the peak charging current of the rectifier circuit might help. When I was building RF level meters a while back I came across an interesting device, the UTD-1000 from Avantek. Inside it, it has a wideband amp with rectifiers that are thermally bonded. The reference rectifiers are used in the output bridge-OP amp circuit. I've attached the circuit of one I made for my TV modulators, and the link for the data sheet. I'm not suggesting that you use the UTD-1000 as the lower frequency limit is too high for your application and they are a very costly part ( I think it cost nearly $100... gulp), but it is worth looking at the way the designers did the rectifier setup in these (I have copied this basic idea for my own level meters with other cheap broadband monolithic amp IC's and separate schottky diodes and it works very very well) You can see on my circuit the two 220k bias resistors that bias the internal diodes. http://pdf.datasheet.live/datasheets...s/UTD-1000.pdf best regards, Hugo. |
18th Nov 2017, 8:19 am | #3 |
Nonode
Join Date: Oct 2008
Location: Warsaw, Poland and Cambridge, UK
Posts: 2,679
|
Re: Minature Coax cable
If they've got SMB connectors on, the cables are almost certainly 50R. If you happen to be passing Warsaw, you're welcome to pop in to my office and check them on my TDR - that will quickly the characteristics of any cables and connectors, though it might not be practical in this instance!
In my experience it's the way the cables are terminated (connectors, solder joints and so on) which cause the trouble. One of the SMA-BNC cables I use often has an intermittent short inside the SMA connector, which causes some very odd effects at high frequencies. I've ordered a new one but it hasn't arrived yet! Chris
__________________
What's going on in the workshop? http://martin-jones.com/ |
18th Nov 2017, 12:18 pm | #4 |
Octode
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,795
|
Re: Minature Coax cable
Hi Hugo and Chris. Hugo..... many thanks for the tips, I will try the series resistor, but not the UTD 1000...... I found a differental opamp circuit using X3 LM324 sections and two diodes in differential mode.... Ok Chris... when I am passing... LOL I guessed that the cables should be 50 R.......... but is that over the entire frequency range or....have they got strange lower frequency characteristics that no one bothers about if they are used in the high UHF region.
I will persevere and when it is working as hoped I will let you know. Incidentally.... I changed the Input coupling caps to 1uF as 0.1uf had a much higher reactance at 3.5 Mhz 1uF certainly helped level that out.
__________________
Should get out more. Regards Wendy G8BZY Last edited by Wendymott; 18th Nov 2017 at 12:20 pm. Reason: Added info |
18th Nov 2017, 4:25 pm | #5 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 14,007
|
Re: Minature Coax cable
When trying to fathom out these sorts of problems I find one of the best ways is to try the effect of a 10dB Pi- or T-section resistive pad for the intended impedance between the source and the cable, or between the cable and the sink.
The 10dB pad essentialy swamps the source/load impedance, so any odd reactances (which may themselves be frequency-dependent) in the source/sink become pretty much irrelevant. MMICs, for example, often do not play nicely at frequencies below their design cutoff: one circuit I played with loved to become an oscillator, frequency dependent on the length of the output coax! I guess there was some sort of strange negative-resistance game going on. |
18th Nov 2017, 4:33 pm | #6 |
Dekatron
Join Date: Jan 2012
Location: Brentwood, Essex, UK.
Posts: 5,347
|
Re: Minature Coax cable
We always used to insert 10dB attenuators between the signal generator and the DUT for the same reason when making precision measurements when I was involved in designing UHF stripline power amplifiers in the 1970's. Happy days when design involved using Smith charts, rulers and compasses. I suppose its all done with computers these days.
|
18th Nov 2017, 5:04 pm | #7 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Minature Coax cable
Is it possible that these cables use copper or silver-plated steel? If so, they may be fine for UHF and microwaves but at lower frequencies with reduced skin effect the steel boosts resistance and inductance thus affecting the characteristic impedance.
|
18th Nov 2017, 6:09 pm | #8 |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,536
|
Re: Minature Coax cable
How does reduced skin effect increase resistance?
Steel may be higher resistivity than silver, but it will still be another conducting path in parallel with the silver.
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
18th Nov 2017, 10:34 pm | #9 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Minature Coax cable
Senior moment. Sorry!
However, the steel core would still mean that the resistance would be higher than a similar copper-cored cable at lower frequencies, and the inductance could be affected. |
19th Nov 2017, 1:04 am | #10 | |
Dekatron
Join Date: Sep 2010
Location: Cheltenham, Gloucestershire, UK.
Posts: 3,077
|
Re: Minature Coax cable
Many years ago I devised a method to characterise transmission lines using near field probes. It was mainly for looking at microstrip and stripline on PCBs but it also works with coax. It works quite well and gives the pF/m and the uH/m which ultimately tells you the velocity factor, characteristic impedance and the dielectric constant used for the transmission line. I've only used this method a handful of times over the years but it is always very useful and always gave very good results.
I keep meaning to formally write up the method as it worked so well. I'm very busy doing some research work at the moment so can't spend much time on stuff like this though. But if anyone is interested to see it I could try and post up a tutorial how to do it. But it might not be for a few weeks. The results are usually very close to reality. But you will need a spectrum analyser with a tracking gen or the equivalent and it needs to cover up to maybe 300MHz. The Rigol analyser should be fine here. For your detector circuit, I wonder if you could just use an AD8307 log amp or are you more interested in SSB peak levels? You won't get much level range from a MAR6 and a diode detector for peak measurements though. Because the MAR6 output isn't terminated properly, your current circuit will have very poor match at the MAR6 input across the whole of the HF/VHF band and this is probably why you are seeing some strange response effects when you add a long coax cable at the input. If the MAR6 input VSWR was (say) 5:1 and the VSWR of the source was (say) 1.5:1 typical then there could be +1.2dB /-1.1dB of mismatch uncertainty. If there is any additional slope or ripple in the sig gen then this could make things even worse. It should be possible to make a diode detector that is extremely flat from LF through to UHF but it might not work that well as a peak detector unless you drive it with a fairly large waveform. If you just want to measure a single tone RF carrier (via a single AF test tone at the modulator input) then maybe choose something like an AD8307 log detector chip as it can detect over a huge range of signal levels. But you will have to make sure you keep (odd) harmonics at a low level as the chip is prone to large errors in reading if harmonics are present. Best to keep them lower than -40dBc but definitely keep them lower than -30dBc. The same applies to the diode detector. In this case even or odd harmonics can give large errors in reading for this detector. eg you could see up to a +/- 6% power error from a -30dBc harmonic for example. Quote:
__________________
Regards, Jeremy G0HZU Last edited by G0HZU_JMR; 19th Nov 2017 at 1:17 am. |
|
19th Nov 2017, 3:06 am | #11 |
Nonode
Join Date: Oct 2010
Location: Tintinara, South Australia, Australia
Posts: 2,340
|
Re: Minature Coax cable
RG285 is indeed 100R impedance.
As Jeremy stated above, the harmonics from the DDS are probably also causing you some headaches. This is why I haven't done anything on a DDS AGC system for my project for a long time. |
19th Nov 2017, 2:09 pm | #12 |
Octode
Join Date: Jun 2013
Location: Huddersfield, West Yorkshire, UK.
Posts: 1,795
|
Re: Minature Coax cable
Hi Peeps... Many thanks for the feedback. The general idea is / was to monitor the peak voltage of a Dual tone generator.. i.e SSB carrier modulated with dual AF tone at 750 Hz and 1.333 Khz. The Diode assy as it stands works well, and is reasonably flat from 14 Mhz upwards to 30 Mhz, it is below 14 Mhz the problems start... I have an alternative circuit to try, actually posted in another thread in this section, re using an VCA822 and diode detector circuit. but ALL comments have been taken onboard..
Re RG 285. What happened.. over the years I have collected various BNC leads with either marked or unmarked coax, a bit of a lottery...being a yorkshire person I loath disposing of useful items, and only now it is evident that sometimes being "careful" is counter productive. Again I assumed the minature coax leads would have been constant.... maybe not. I really dont know why I am being so exacting with this generator, as it is only for "in house use", and really a "ball park" source, not a standard. However it would be nice to be flat to say +/- 1db.
__________________
Should get out more. Regards Wendy G8BZY |
19th Nov 2017, 6:21 pm | #13 |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,536
|
Re: Minature Coax cable
Flat to +/-1dB is actually pretty good going!
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |