Q-meter questions

[Argus25]

Originally Posted by G0HZU_JMR

Yes, a typical x10 scope probe will often prove to be unsuitable as a high impedance sniffer.

At RF frequencies, the equivalent shunt resistance of a x10 scope probe will drop quite quickly. By 5 or 6MHz it could be as low as 100,000 ohms in parallel with about 12pF. By 20MHz that will drop to maybe 8,000 ohms in parallel with 12pF. So this is not good at all and will usually damp a tuned circuit.

This is why in repairing/adjusting some RF gear I use the x10 probe by clipping it onto the insulation of a wire, rather than the actual wire and winding up the scope gain. It creates a fraction of a pF gimmick coupling cap forming a divider with the probe's capacitance, enabling the waveform (with now meaningless amplitude information) to be observed without significantly altering or de-tuning the circuit , very useful for peaking stages in radios, TV etc.

[regenfreak]

Yes, a typical x10 scope probe will often prove to be unsuitable as a high impedance sniffer.

This is exactly what the graph in the Digikey link shows in my previous post and why I am thinking about the cheap active probe.


https://www.digikey.co.uk/en/article...loscope-probes


I think the Q of the gimmick cap is important and it has to be fairly high to minimise dielectric losses . In the cheap Ukrainian active probe article and the DIY FET-amplifier article, you will see the tips of the active probe or FET amplifier have to be carefully made and designed:

http://www.crystal-radio.eu/fetamp/enfetamp.htm

0.3pF input copper plate capacitance would have to be estimated using empirical equations based on the dielectric properties. I am not sure any capacitance meter can measure this.

[Radio Wrangler]

Scope probes aren't *quite* as bad as they're being painted. A lot of the drop in impedance at higher frequency is capacitive reactance. This will pull the tuning of any tank it is probing, only the resistive component presented will damp the Q. The resistive component also falls, so the effect is still significant.

David

[David G4EBT]

For what it's worth, I've attached another constructional article on a 'Q' Meter. Namely, the 'Cymar' which appeared in Radio & Electronics World in January 1984. Despite the design being 35 years old, it doesn't require any exotic obsolete components.

I always thought that R&EW was published by AMBIT - later Cirkit, who sold components and kits back in the day when people still liked to build things rather than to buy them. However, according to the americaradiohistory site, it states that R&EW was a successor to Radio Constructor, which ceased publication in 1981. ( R&EW ran from 1981 - '89, and though there is a section for it on the ARH site next to Radio Constructor, there doesn't appear to be any scanned issues as yet).

Hope that might be of interest.

[G0HZU_JMR]

There do seem to be a few similarities between some of the homebrew designs. The Cymar design uses a test source with a very high source impedance which we know isn't a good idea.

It also compounds this by using a common source JFET buffer amplifier ahead of the detector. This is definitely not a good idea as the input (Rp shunt) resistance of the common source JFET in that circuit will drop alarmingly even by 2MHz. Rp could be as low as 50k ohm in parallel with 10pF by 2MHz and it could be less than 5k ohm in parallel with 8pF by 20MHz. That's even worse than a x10 scope probe.

So this design should definitely be avoided unless the aim is to make something that allows crude comparative measurements of Q. In other words, this design is a toy at best.

[Philips210]

Thank you David for the details on the 'Cymar' Q-meter. On the ARH website, a number of issues of R&EW have been added recently: https://www.americanradiohistory.com...nics-World.htm

Regards,
Symon.

[David G4EBT]

Originally Posted by Philips210

Thank you David for the details on the 'Cymar' Q-meter. On the ARH website, a number of issues of R&EW have been added recently: https://www.americanradiohistory.com...nics-World.htm

Regards,
Symon.

Thanks Symon - that's good news.

What a wonderful resource.

[trh01uk]

There's someone advertising a Wayne Kerr B801 admittance bridge on the vmars email list. These are the details of the post:

The following Wayne Kerr Admittance Bridge set of units is going to the Tip; unless somebody is curious about it.

Vintage:- is about 50 !! years old.

VHF Admittance Bridge; type B801. [I think this is based on Ratio Transformers with switched selection of 'Taps'.]

[allows 'Y' parameter measurements used with control unit below.]

Bridge Source; type S1618. [5MHz to 100MHz.]

Control Unit; type C.U. 681. [Provides bias volt and current setting for active devices.]

Transistor Adaptor; type Q801A.

Standards; type Q761 [components in polished wood box].

Anyone who wants to follow this up and is not on the vmars list - please PM me.


Richard

[G0HZU_JMR]

Originally Posted by Radio Wrangler

Scope probes aren't *quite* as bad as they're being painted. A lot of the drop in impedance at higher frequency is capacitive reactance. This will pull the tuning of any tank it is probing, only the resistive component presented will damp the Q. The resistive component also falls, so the effect is still significant.
David

In my experience the loading effect of a typical cheapo x10 scope probe can be modelled by the tip compensation (13pF?) in series with the capacitance of the cable (120pF?) in series with a loss resistance for the coaxial transmission line. I usually find the loss resistance is around 50 or 60 ohms. This model simplifies to about 12pF in series with 60R.

When this is transformed and plotted as Rp and Cp it gives the equivalent parallel (damping) resistance and the parallel capacitance vs frequency.
I think I've posted up this type of plot before but it might be useful to see it again on this thread. It shows how quickly the Rp value drops as the frequency increases. The plot is just for a model but a typical scope probe will look very similar. I've seen some x10 probes where the equivalent model is 90R in series with 11pF but usually it's around 50 or 60R in series with 12pF and this basic model holds true across maybe 1MHz to 100MHz or so. So the plot below should be quite useful I think.

[Radio Wrangler]

I use somewhat better probes than the usual cheapies, and I avoid the switched x1/x10 jobbies, their loading is still a problem for any RF work.

Sealectro did some rather useful resistor tips for SMC cables in 500 anf 5k Ohm flavours, these are a lot more useful at moderate RF. Tek do a'passive' probe which is also a resistor ahead of normal 50 Ohm coax. The price isn't good for what it is!

Otherwise I break out the HP 500MHz active probes.

David

[Al (astral highway)]

Originally Posted by regenfreak

Al I gather you are the expert on tesla coils as I have seen your threads ... I don't like handling MOTs, they are one-touch killers and they make neon sign transformers and flyback transformers rather "safe"

Exactly, they are one-touch lethal. I enjoyed my excursion into them for a while, but I had an ahah-moment one day and got rid. I will never encourage/endorse a build with one now. Same goes for the newer type with a high-powered inverter on board. They're potentially even worse, as silent (unless you can hear the supersonic component at 25KHz or so).

Originally Posted by regenfreak

How do you measure your Q? I know some coilers like to use ring-down method...or ring up and ring down

To be honest, I calculate it before the build, and then I wait to see the results.

Having said that, I have looked all around for decent research and I found it. The take-away is that Tesla secondaries tend to have a Q between 200 (statistical mode) and 280. I've heard some claim Q's of 400 but this may be exceptional.

This is from the IEEE's transactions on plasma science. Source: 'Optimising the secondary coil of a Tesla transformer to improve spectral purity 42(i) pp143.148

Authors: Raven, RM; Smith, IR; Novac, BM. Published under the creative commons license, so extensive citation is ok with acknowledgement.

The paper addresses Tesla coil usage in non-recreational areas of usage including EHT generation for physics (especially plasma) research environments and so on.

If the complex impedance of the coil is measured by a vector network analyser (VNA)...measuring this impedance at resonance and then finding the frequencies at which the inductive reactance and the capacitative reactance are equal in magnitude to eachother enable half-power points to be found for the fundamental and also higher mode resonances, denoted by fn.

This response can be measured, since the lower (fl) and upper (fu) frequencies, corresponding to a halving of the input power, can be determined.

Then Q is given by:

Q=fn/fu-fl

This is nice. I don't have a VNA at the moment, although forum member MerlinMaxwell kindly sent me a mini-one. I just haven't got round to getting onto a Windows environment yet.

Data published by this research group shows, as I say, the commonest value of Q to be 200. It peaks for secondary coils self-resonant at 630KHz, where it may reach 280, and tapers markedly for coils resonant at 800KHz or higher.

What do you make of all that?

[regenfreak]

Data published by this research group shows, as I say, the commonest value of Q to be 200. It peaks for secondary coils self-resonant at 630KHz, where it may reach 280, and tapers markedly for coils resonant at 800KHz or higher.

What do you make of all that?

My biggest 1.7kW spark-gap tesla coil has the predicted secondary Q of 180 at resonance frequency of 139KHz using two large spun aluminium toroids of 20" and 11.5" sitting on top of each other as the top load.

I calculated Q using Javatc 3D software. I have tried three different combinations of secondary coils and two primary configurations with different primary surge impedance, they all have predicted values of Q around 180-200. I have never been able to measure the actual Q.

http://www.classictesla.com/java/javatc/javatc.html

But I will be able to do once I receive the new toy Siglent SDS1202X 200 MHz scope tomorrow. I have receieved my Poor Man's active probe from Ukraine today. It is harder to measure Q with tesla coils because any human body or walls near the coil will affect the results. The same is true with the measurement of resonance frequency that I have to model the spark loading with a wire during physical measurement and software modelling with Javatc.

I think Q is not as critical as coupling coefficient k in most cases becuase you cannot change it once you wind the coils. I make my tesla coil having adjustable coupling coefficient k (very fine adjustment of secondary coil using screw mounted riser platform).

This is nice. I don't have a VNA at the moment, although forum member MerlinMaxwell kindly sent me a mini-one. I just haven't got round to getting onto a Windows environment yet.

I am not qualified to use a VNA until I understand Maxwell equations, transmission line theory and the Ying-Yang Smith chart, ... The cheap VNA starts from £430 on ebay. I tried to understand Maxwell equations and antenna theory when I first started my first ever homebrew radio project; a spark gap transmitter and a silver-nickel filings coherer detector six months ago; I didnt really understand the maths parts of the Maxwell equations and dipole antenna theory...I understand the ideas though.

Exactly, they are one-touch lethal. I enjoyed my excursion into them for a while, but I had an ahah-moment one day and got rid. I will never encourage/endorse a build with one now.

The number of hobbyists who are killed by MOTs annually and globally is unknown. You can walk away to tell the tales with nasty burns or deep nerve damages with flyback or neon sign transformers. The next deadly ones up the scale are the potential and x-ray transformers
I think it is healthy to feel scare when one is handling high voltage because I develop the respect for it. I have never got zapped once since I started designing and building tesla coils.

[regenfreak]

To be able to understand what are the limitations of the oscilloscope probe and coaxial cable, I will really have to grind through this Feng Shui chart and stomach the abstract concepts; the Ying Yang of L network matching with different topologies and frequency response:

http://highfreqelec.summittechmedia....E0306_Rhea.pdf

The "Q of load" is discussed in part 2. :

http://www.highfreqelec.summittechme...Rhea-part2.pdf

Well more on my reading lists. It is difficult stuff.


Figure 5 and 6 are very interesting!

[G0HZU_JMR]

As I said earlier, I have no experience with Tesla coils. However, if you just want to know the Q of the Tesla resonator when it is stood vertically on a base then I'd have thought you could use the E and H field method. But this is just a guess.

Ideally, you would want to use a swept system like a (scalar) network analyser or a spectrum analyser with a tracking gen. But you could also use something cheaper like the Analog Discovery 2. The idea is that you transmit via the port 1 of the analyser with the H field probe and you receive with the E field probe on port 2.

Move each probe closer and closer to the coil until you get a peak reading on the swept display and try and keep the probes far away enough that you get a steady reading but the bandwidth doesn't widen (through remote loading of the Tesla coil). Then just measure the 3dB bandwidth of the peak on the analyser and you can work out the Q. I use this method a lot for LC resonators or helical resonators. The H field probe couples a tiny amount of energy into the coil and the energy then sloshes back and forth inside the resonator between E and H fields. The E field probe can then pick up the E field part of all this. As I said earlier, I wrote some software that runs on my little netbook and this controls the VNA and it measures the Q in real time using active markers on the VNA. So once connected up, it can measure Q in real time. It will also measure the loaded Q if you bring anything near the coil that loads it. eg another smaller (primary?) coil.

This method does away with source impedance error and also the detector error because the coupling to the system is remote and very light. But you do need a sensitive analyser as the signals returning will be in uV. With a coil as big as a Tesla coil the E and H field probes could probably be a couple of feet away and still get a solid measurement.

But this might not be the goal you are seeking. I'm just guessing that you want to know the Q of the Tesla coil when it is acting as a resonator?

[Al (astral highway)]

Jeremy , did you read the method in my post no. 91 up there , from a published research group /IEEE (not recreational Tesla coil use , for physics research applications)

Cheers

[Radio Wrangler]

Whatever method you use, the coupling into and out of the resonator under test has to be VERY light. Trad Q meters need a tiny series impedance to inject signal and a very high resistance on their metering side. These are two ways of achieving light coupling electrically. Jeremy's field probes have to be used a good distance from the resonator for light coupling.

The aim is that the losses of the input and output couples have to be much smaller than the losses of a tank of the highest Q the instrument is to be capable of measuring with reasonable accuracy.

If the couple losses are known, indicated Q figures can be corrected to an extent, but this multiplies errors just at the end of the scale where things get interesting. Murphy!


David

[G0HZU_JMR]

Originally Posted by astral highway

Jeremy , did you read the method in my post no. 91 up there , from a published research group /IEEE (not recreational Tesla coil use , for physics research applications)
Cheers

Hi Al, yes I had a quick look at this bit...

If the complex impedance of the coil is measured by a vector network analyser (VNA)...measuring this impedance at resonance and then finding the frequencies at which the inductive reactance and the capacitative
reactance are equal in magnitude to eachother enable half-power points to be found for the fundamental and also higher mode resonances, denoted by fn. This response can be measured, since the lower (fl) and upper (fu) frequencies, corresponding to a halving of the input power, can be determined. Then Q is given by:

Q=fn/fu-fl

That is classic stuff but I'm not sure how to apply it to a Tesla coil. Normally, that method would be used for series resonant circuits and especially those where you can't easily measure the inductance or the capacitance but you can measure the ESR at resonance easily and also easily measure the +/- reactance points that have the same magnitude as the ESR. This should be at the +/- 45degree points of the zin measurement. Zero degrees would mean it was all resistive, +/-90degrees would mean it was all purely reactive and +/-45degrees is the sweet spot referred to as the half power points in your quote above. So I would have thought this method is more suited to measuring the Q of something like a regular RF crystal. This could be done with a vector voltmeter or a decent VNA using this method.

But the Tesla coil is a parallel resonant system. A revised method would be to look at the right side of the smith chart. In theory at least you could find the points where the +/- imaginary value matches the parallel resistance Rp (at resonance). This would give the 3dB BW. But I don't think a regular VNA can do this because the impedances are so high. So I think I've missed what the research group means here.

[regenfreak]

Thanks Guys. I will stick to the ring-down method for measuring Q in tesla coils. It is the simplest method and requires just a scope and a function generator. The other Q measurement technique is to use frequency sweep 3db method using an "antenna" dangled above the tesla coil and connected directly to the scope. I will leave the VNA and Smith charts to the experts.


I will also try to use very weak inductive or capacitive coupling method using the cheap active probe to measure Q. The active probe uses dual-gate Mosfet (like a tetrode) with a gate capacitance of about 2pF. As a hobbyist, I don't have deep pockets like you pro guys here. The cheap Ukrainian active probe from ebay has very small capacitive loading on the circuit (0.75pF) and high input impedance at the typical resonance frequency of tesla coils (100kHz to 450kHz). With frequency less 1MHz, the poor man's active probe is a very effective RF probe.

This video (Mr Carlson) demonstrates the value of cheap DIY active probe (different design but similar idea):

https://www.youtube.com/watch?v=NlbC-cMIBTY

This guy did his university thesis on the evaluation of DIY active probe:

https://cdn.instructables.com/ORIG/F...ZPIPJTFO82.pdf

The passive 1:10 probe tends to load down the Q and detune the tank circuit with its large capacitance and low impedance at higher frequency. The passive 1:100 probe has too much attenuation and the amplifier noise or vertical sensitivity of the scope will become a big issue.

The Q for the primary coil of a tesla coil is usually extremely high (never needed to be predicted or measured) because it is made of very thick copper pipes in order to minimise the skin effect. My large tesla coil has about 800-1000 amps current through the primary coil.

The general rule is that Q for the secondary coil should high but there is no advantage to make it very high because once the sparks hit as the ground objects, the Q will be loaded down anyway. To improve the performance of a tesla coil with a given input power, it is better to increase the surge impedance of the secondary coil by increasing the secondary inductance in order to reduce the massive current through the primary coils. The higher the primary current, the more losses through the spark gaps. Also it helps to use large tank capacitance to move away from the resonance rise of the neon sign transformers and use the biggest topload to create the fattest and brightest sparks.

This describes two Q measurement techniques for tesla coil:

http://www.teslathon.de/stefan/tc/tech4.htm

PS I always deliberately detune the tesla coil so that the resonance frequency coincides with the condition during the longest spark ground strike. The spark acts like stray capacitance and lowers the resonance frequency significantly.

[G0HZU_JMR]

This video (Mr Carlson) demonstrates the value of cheap DIY active probe (different design but similar idea):

https://www.youtube.com/watch?v=NlbC-cMIBTY

Sorry to be a neghead (again) but that probe design is very strange and I don't recommend anyone makes it unless they want to look at really small signals with a limited dynamic range. The overall design is not good. Best to leave it at that.

The cheapo Ukranian probe is going to be a better bet I think.

[G0HZU_JMR]

The other Q measurement technique is to use frequency sweep 3db method using an "antenna" dangled above the tesla coil and connected directly to the scope.

Yes, that's very similar to the swept method with the E and H field probes. In the case of a vertical resonator I generally feed the coil under test near the base with the H field probe and then have the E field probe near the top end of the coil.

Presumably your new scope will have an FFT mode and if you select it you should be able to configure it to see quite small signals just like a spectrum analyser. You should be able to get down into the uV region like this although you would be best to select the 'flat top' window function if you want the best amplitude response behaviour as you sweep a signal with a sig gen and watch it on the scope.