InsertionVSWR of Chinese 1:9 balun module #3 – correcting Z using e-delay

The article InsertionVSWR of Chinese 1:9 balun module #2 gave a plot of R,X where the reference plane was at the compensation capacitor on the module. Recall that the transformer was actually 1:4.

Ideally you might OSL calibrate the fixture at that point, or at the transformer terminals to obtain a good picture of the transformer response, but it is not convenient to be desoldering the transformer or cutting tracks, so there is a fairly good alternative at the frequencies being discussed (1-30MHz).

Above is an annotated pic of the test setup. The NanoVNA is calibrated at the Port 1 connector (in fact the outboard end of the SMA savers that are never removed), and the desired reference plane is some 32mm away of transmission line of segments of uncertain Zo and loss.

Firstly, the loss will be very low, so we can simply deal with the time delay that the fixture introduces.

The NanoVNA-H4 with DiSlord v1.1.0 firmware has a facility for port extension (though it is not called that, it is called e-delay), as does the NanoVNA-App software used. In both cases, the value of e-delay converted to phase is subtracted from the s11 measurement at each frequency, and the result is correction for the port extension delay.

Note that this:

• does not compensate amplitude change, so it is only accurate where loss is very low; and
• assumes that the port extension has Zo=50+j0Ω though mixed Zo does not cause significant errors if the transmission line sections are very short.

Let’s apply a temporary short circuit the the compensation capacitor at the reference point.

We want X=0, the rising X is due to the transmission line transformation within the test fixture, and that transformation will render any uncompensated impedance values incorrect (though the ReturnLoss and VSWR will be correct).

Lets look at the phase of s11 for the same case, remembering that the phase of the short circuit should be 180°.

s11 lags 180° by 4.6° @ 30MHz, we can calculate the time equivalent: $$t=\frac{\phi}{2 \pi f}=\frac{4.6 \pi}{360 \pi f}=\frac{4.6}{360 f}=426 \text{e-9s}$$, so 426ns is a good starting point for e-delay.

Ok, returning to the R,X chart and adjusting e-delay around 426ns for best X plot…

Above, 430ns is pretty acceptable.

Having backed out the delay of the fixture, lets look at R,X.

Above, R,X with and without the e-delay adjustment. The difference is quite small in this case (the lower R and X @ 30MHz are the e-delay corrected values).

Applying the calibration short circuit

In the experiment above, an X-ACTO #16 knife was used to bridge both sides of the capacitor mentioned. The same technique can be used to bond a track to an adjacent ground track (the knife will cut through the solder mask).

A selection of short circuit coax adapters, M & F, can be very handy when making measurements.

For example, to measure an antenna with a UHF female connector, a short cable with adapters SMA(M) for the VNA to UHF(M) for the antenna can be calibrated with a UHF(F) SC (made simply with a panel jack and the back side shorted with several copper conductors soldered as close to the insulator as possible).

Even if the place where you can conveniently apply the reference SC is not exactly where you want it, if the further extension is something you can characterise well (like a measurable length of known 50Ω line with known VF) you can further tweak the e-delay setting.

For example, to measure an antenna with a UHF female connector on a 400mm pigtail of Belden 8267 (RG213) from its actual feed point, a short cable with adapters SMA(M) for the VNA to UHF(M) for the antenna can be calibrated with a UHF(F) SC. Having found the appropriate e-delay to the applied short circuit, calculate the one-way delay of the coax pigtail. Using TLLC we get 2022ps, so the round trip delay is 4044ps. Now add that 4044ps to the e-delay to the applied short to get the appropriate e-delay to the actual feed point.

Summary

The NanoVNA-H4 with DiSlord v1.1.0 firmware has a facility for port extension (though it is not called that, it is called e-delay), as does the NanoVNA-App software used here. In both cases, the value of e-delay converted to phase is subtracted from the s11 measurement at each frequency, and the result is correction for the port extension delay.

Note that this:

• does not compensate amplitude change, so it is only accurate where loss is very low; and
• assumes that the port extension has Zo=50+j0Ω though mixed Zo does not cause significant errors if the transmission line sections are low loss, close to Zo=50+j0Ω, and very short.

The technique may provide a means of moving the reference plane with acceptable accuracy.

ESP WiFi relay project – preview

The impetus behind the project is a remotely WiFi controllable relay for reset function in a remote controlled ham station.

The information presented here applies to development v0.1.

Features:

• support typical multi channel relay boards;
• ESP8266 and ESP32 firmware versions;
• WiFi credentials programmable via a captive web interface;
• DHCP or static IP;
• nMDNS responder;
• flexible configuration stored as json file in on-board LittleFS file system;
• optional authentication to secure remote access.

A variety of integrated relay boards

Above, a Yunshan relay, not recommended as RESET pin is tied high. This one was binned, quite a waste of money. It exemplifies a common problem in that the RESET line is often not available on a header pin, this one is worse in having tied RESET high and no header pin for EN. Continue reading ESP WiFi relay project – preview

This article documents an InsertionVSWR test of another cheap Chinese 1-9 balun purchased for less than <$5 on eBay (shipped). Above is the advertising pic of the 1-9 balun, it would seem to be a clone of the Noelec 1-9 balun. The balun is a compensated voltage balun with the secondary centre tap grounded for these measurements. Continue reading InsertionVSWR of Chinese 1:9 balun module #2 A 1:4 RF transformer for measurements – based on Noelec 1:9 balun assembly The Noelec 1:9 balun (or perhaps Chinese knock off) is available quite cheaply on eBay (<$5) and provides a good hardware base for a 1:4 version.

Above is a modified device with the original transformer replaced with a Mini-circuits TC4-1TG2+ 1:4 transformer. The replacement is not exactly the same pads, but it is sufficiently compatible to install easily.

The loose parts are OSL calibration parts using 2.54mm pitch header pins with the middle pin removed. The resistor comprises two 100Ω 1% 1210 SMD resistors soldered back to back. (The 1% resistor code 1000 is for 100 * 10^0.)

The most notable departure from ideal of these small transformers is leakage inductance of 30nH give or take. Continue reading A 1:4 RF transformer for measurements – based on Noelec 1:9 balun assembly

NanoVNA-H4 v4.3 – broken SMA connector

It has been my experience over more than 50 years that accurate measurements using SMA connectors requires that they be torqued to a consistent and adequate torque. Specifications for brass threads commonly runs around 0.6Nm, and I have a torque wrench calibrated for that torque which I use whenever the connectors are to be properly tightened.

In my NanoVNA-H v3.3, I reinforced the SMA connectors because of a sense that to tighten them to 0.6Nm (5.3inlb) caused the board to flex and over time might crack the tracks (Strength of reinforcement of nanoVNA-H connectors).

Joe Q Smith on NanoVNA SMA connectors

Joe Q Smith has an interesting video where he tests some Chinese SMA connectors to destruction, worth watching: NanoVNA Torquing SMA Connectors . In his tests, he needs upwards of 2Nm to damage the cheap Chinese end launch PCB connectors, more than three times the torque I use.

The new NanoVNA-H4

About 6 weeks ago, I took delivery of a new NanoVNA-H4, which as far as I can tell, is a ‘genuine’ Hugyen product.

I did not reinforce the connectors on my new NanoVNA-H4 because there was not the needed clearance for a similar adequate brass bar inside the case.

So, having used a torque wrench at 0.6Nm in the few weeks that I have owned the NanoVNA-H4, the connector on Port 1 failed.

Above, the failed connector has rotated, shearing two of the support pins and tearing the other two off the tracks. Continue reading NanoVNA-H4 v4.3 – broken SMA connector

NanoVNA-H4 v4.3 – improving the SD card slot

One of the shortcomings of the NanoVNA-H4 v4.3 is that it is quite easy to drop the SD card inside the case when trying to insert it. Experience is that this is really easy to do in difficult field situations or poor lighting / visibility.

This could have been prevented by better design of the moulded case.

This article describes a simple modification to make it more difficult to miss the card slot.

Above, the modification is a small block of plastic that covers part of the aperture moulded into the case, it projects 1mm into the aperture. Continue reading NanoVNA-H4 v4.3 – improving the SD card slot

Transformers and flux density

Lots of online discussions on ferrite cored transformer losses might make you think that the core material is in a path in series with the transferred power and that it acts to some extent like an attenuator.

That sort of thinking betrays a lack of understanding of how a transformer works.

If you take a good 50/60Hz 1:1 power transformer, assume no losses, no flux leakage, and ignoring distributed capacitance, you might ask: Continue reading Transformers and flux density

Selfevidently

An online expert held forth on the design of ferrite chokes and transformers, and to quote one paragraph:

Equally selfevidently we don’t want ANY real part of the reactance in a transformer and, for a practical transformer, we want the self inductance on each side (primary and secondary) to be at least j10*R(Load or Source) and the coupling to be as close to 100% from primary to secondary. It is the real part that heats up transformers a LOT and, since ALL of the current is seen by the ferrite in a transformer, not just the part that got reflected back on the outside of the coax in a choke, losses are abos-posilutely-undubiously NOT desired and the u”R needs to be as close to zero as we can get at the designed frequency for minimum loss and minimum power dissipation.

Setting aside the hyperbole and the wooly thinking, let’s drill down on u”R needs to be as close to zero as we can get at the designed frequency for minimum loss and minimum power dissipation.

It is a pretty general statement without really specific quantities, needs to be as close to zero as we can get and minimum loss and minimum power dissipation does not give useful guidance of acceptable values of µ”, and may even impart the impression that the following chart is for material that is not suitable above perhaps 200kHz, if that.

Above, µ” is greater than 10 above about 200kHz, greater than 100 from about 2 to 100MHz. Is this what the quote condemns? Continue reading Selfevidently