I am hoping for some advice on this so I can finish my QMX. I found another thread on this, but it didn’t help me. Specifically, should both T507 10 turn windings be in the same direction , either both CW or both CCW, when viewed from the same side of the binocular core?

Tnx es 73,
Randy, KS4L

Not sure it matters. I followed the manual which has you wind them
with the same sense.

-mike/w1mt

I believe that it does matter.? Reference this article:


Note that the transformers are marked with a sense (or polarization) dot.? The math would be off if one of these were opposite (the windings not sensed the same).?

This is my interpretation.? I could be in error and would welcome any feedback.
73
Evan
AC9TU

The following shows the proper winding of a Stockton bridge on a binocular core.? ?See pages 10-12.



Walt

MFC: Motion, Flux, Current the "law" of generators or something like that recalling basic electricity Navy school. "Right hand rule"? fingers show direction of current through the coil, thumb shows direction of flux.

Reversing either direction of current or winding of coil wil reverse the flux direction.

73
Karl
KI4ZUQ

On Tue, Aug 1, 2023 at 10:28 PM, Randy wrote:

Does the "handedness" of the 10 turn windings on each end of T507 matter? I can't be sure how they are done on the photo on p. 40 of the assembly manual.

?

Tnx es 73,

Randy, KS4L

If one winding is wound different to what the designer intended, then the "forward" and "reverse" sensing of the reflectometer will be reversed.? Hans could provide a configuration parameter to cope with this eventuality.? We won't know if this is a problem until he enables the reflectometer functionality in the firmware, but there may be a few people who have done it different to the way that Hans has wound his prototype...
?
--
Peter Lee
G3SPL

I agree. We have "right handedness" of every coil done so far and we have Walt's link:

On Sat, Aug 5, 2023 at 01:55 AM, Karl wrote:

Walt
And they will be different. The schematic may show that Hans reversed one coil to make the bridge work (or to make the PCB connections work) BUT without knowing for certain we may face a do over.

I'm not crazy enough to want a "do over" on this one!
Hans! What say ye?

73
Karl
KI4ZUQ

I think the simplest option to prevent "do overs" would be for Hans to provide a configuration option to reverse the reflectometer "forward" and "reverse" selection which would be simple to do in firmware.? Alternatively, he could check for which sample is larger than the other, and automatically choose that as the "forward" one.? That, however, would be problematical if, for example, the output was infinite SWR (e.g. open circuit) when slight differences in Schottky diodes etc could make the "reverse" output slightly bigger than the "forward".??
If the configuration option were chosen, it could be accomplished by connecting a dummy load, or a known well matched antenna, and doing a "setup reflectometer" action where the firmware would select the higher output sensor as "forward".
?
--
Peter Lee
G3SPL

I think that you may be overthinking this; the natural way is for the windings to go from a hole in the board, through the centre of the core, back round the outside, through the centre until there are 10 passes through the centre, then the wire come out of the core and into the hole in the board on the other side.

To wind the core the other way you would have to go from the board and round the outside of the core before going back through the centre. If it was wound like this it would be obvious, there would be loose wire everywhere.

Both transformers have to be the same. so their outputs are in phase.

If you really think there's something wrong with the design then don't install it.? There's no support in software for it at present.? Fit a wire in place of the top single turn.? Then wind and fit T507 when people have demonstrated that it's correct.

Chris, G5CTH

Asked and answered already in "QMX documentation":

开云体育

Assuming you've been sensible enough to design the PCB to make the natural way correct :)

I didn't think it would be difficult to provide more of the diagnostics you must have needed when developing this.? Not that surprised it's already there.

I make the two VSWRs 1.06 and 35:

vswr = (fwd + rev) / (fwd - rev);

It's a ratio so should work with the raw ADC values.

I can't see a reason to provide a swap function.

Chris, G5CTH

Hi,
perhaps one of this docs can explain why the winding direction in this kind of a VSWR bridge is important.
FW is very important here but HW is too!


73 Bojan S53DZ

开云体育

On Sat, Aug 5, 2023 at 05:11 PM, Chris wrote:

make the two VSWRs 1.06 and 35:

vswr = (fwd + rev) / (fwd - rev);

It's a ratio so should work with the raw ADC values.

The SWR calculation using just raw ADC values is not adequate, I am afraid.? This is because we are using low powers where the effect of the forward voltage characteristic of the diodes cannot be ignored, particularly for the reverse sensor where (in the well matched case) the reverse sample will be very much lower than the forward.

Mike, G8GYW, explains this well in his GitHub article (URL https://g8gyw.github.io/). He worked out that the diode characteristic is of the form y = ax² + bx, with the parameters of a = 1.0 and b = 0.75.

However, for the tandem bridge I built, perhaps because I was using different diodes, I found that the parameters needed were a = 1.02 and b = 0.42.? These parameters were found using an Excel spreadsheet to plot reflectometer readings with 7 different supply voltages (to a QDX) and measuring the power output into a dummy load.? These readings were plotted on a chart together with the y = ax² + bx curve and I varied the a and b parameters until I got a good match of the curve with the measured spot values.

Other useful info on the tandem bridge reflectometer are G4ZNQ's article in "Sprat" (URL https://www.sm7ucz.se/Meters/Stockton_pwr_meter.pdf) and a detailed description at K5JCA (URL https://k6jca.blogspot.com/2015/01/notes-on-directional-couplers-for-hf.html).?

--

Peter Lee

G3SPL

Hans,

If you reverse the values of FWR and REV, you wind up with the same ratio with a different sign.?
Assume FWR = 100,? REV =10, SWR = 110/90 = 1.22
? ? ? ? ? ? ? FWR = 10,? REV = 100,? SWR=110/-90 = -1.22

As long as you take the absolute value, you are OK.? It does matter if you do not take the absolute value.

The above is with the assumption the values will stay the same.? I am not convinced they will, as now the phase will be different between the current and voltage used to measure power.

So at least the ability to change the FWR and REV signals would be required.

73
Evan
AC9TU
?

开云体育

Hi Peter, yes the non linearity of the diodes makes this VSWR value approximate.? But for the purpose of tuning an antenna the minimum value will be at the same tune point.? And for protecting the finals the voltages calculated will be close enough.

And I think it's worth pointing out that the equation you give, y = ax² + bx? gives the power, not a voltage.? This means that the Vswr calculation will need changing.

AIUI the mode by which the BS170s fail if the VSWR is high is by exceeding the voltage limit; is the relevant voltage a function of the sum of the forward and reflected voltages?? Could VFWD + VREF be used to limit the maximum PA voltage to something that's safe?? It may need to be different for the 9V and 12V versions.

Chris, G5CTH