Tuesday, April 28, 2020

Autotuner installed into mcHF

After a three hour surgical procedure my mcHF now has an internal ATU installed. The process was easy enough and would be do-able by just about anyone who's built a few kits and knows how to solder and desolder competently. Lacking this experience, I would not recommend it.

My mcHF, field-stripped. Not shown is all the hardware that holds it together

The ATU is compatible with a variety of mcHF versions as well as the pirated Chinese clones. My particular version, the 0.6.3 mcHF, requires the removal of the RIT encoder (E3) so that it can be replaced with an encoder that also incorporates a pushbutton to actuate the ATU function. An LED also has to be removed and then replaced with a 2-color (three leads) LED, and a static discharge component has to be relocated from the mcHF to the ATU board. Lastly, the wiring harness has to be connected to various pins on the mcHF.

It is a bit involved and I took plenty of photos along the way so I'd know how to fit everything back together, especially all the small hardware holding everything in place.

I've perfected my part-removal technique over the years and recommend it to those who need to remove a non-functioning part from a PC board. This is my patented technique and, each time you make use of it, you have to mentally send me a beer:

I first cut the components away with diagonal cutters. Then, with a wooden toothpick ready, heat up the hole while simultaneously pushing the original part's remnant on through to the other side. You'll probably have to whittle down the diameter of the toothpick first as they are usually too thick to go through the circuit board's holes that you're attempting to clear.

Once this is done you're left with a clear, solder-free hole, ready for the new component. No solder wick is necessary and - this is the best part - far less heat is applied to the board/traces.

Photo time:

Removing encoder E3, and then LD2

New encoder and 2-color LED installed and wired

DS1 has to be removed from the mcHF's RF board...

...and re-installed onto the ATU circuit board

A pair of wires inserted into DS1's prior QTH

RF board completed with ATU mounted, harness wired

The stand-offs that support the ATU may seem like a flimsy mount - they are not. ATU is rock solid.

Close-up of mounting, wire routing and DS1 relocation

The rest of the installation was mostly point-to-point wiring (and the preparation of one end of each wire) and the physical mounting of the ATU onto the mcHF's RF board via four soldered stand-offs. One hole to drill (1/16th of an inch) and a two-pin header to install and that was it - time to reassemble.

The autotuner can be engaged or disengaged from operation. After cycling through these states, I tested it on a 50-ohm dummy load and then on my home antennas, including tuning my triband Yagi on 12 and 17 meters with no problem.

After that, I used my antenna analyzer to detune my 30-40-60-80 meter dipole until the SWR was 3:1. Then I hooked up the newly outfitted mcHF and let the autotuner perform its magic. No problem on any band with a 3:1 mismatch, therefore the tuner satisfies my requirements for the antennas I'd be most likely to use while portable.

One thing I don't like about the ATU and that you should consider before buying one, if you're a mcHF owner:

The ATU does not have memories; each tune cycle is a full tune from Square 1, even for a frequency you've been on before. If you're a KX2 or KX3 owner (or any other of a wide variety of rigs), you know that if you return to 20m after having been on (and performing a tune) on 17m, that return to 20m will cause the ATU to assume its previous values. Similarly, if you initiate a tune cycle on a previously tuned frequency, the ATU in those rigs will tune quickly - almost instantaneously, thanks to memories contained within those ATU's. Not so with this ATU for the mcHF. It is not interconnected with its host transceiver the way most other brand-specific tuners are. It is best thought of as a stand-alone ATU that just happens to be housed internally in the mcHF.

And one criticism of the mcHF has to do with the rig itself - not the tuner:

The SWR meter built into the mcHF is unreliable in that it can't be adjusted to give a 1:1 indication on all bands into a perfect 50-ohm load. On some bands, even though the match is perfect, it may indicate 1.8:1 or similar. If you adjust it on that band, it will then be mis-adjusted on another band. This is with the mcHF in its original state, with no ATU installed.

Unlike with the Elecraft KX2 or KX3 rigs with the internal tuner option, there is no digital read-out upon completion of a tuning cycle to tell you what your new SWR is - you just have to use the mcHF's internal meter and know (for example) that 1.5:1 on 20 meters is really 1:1.


PE4BAS, Bas said...

John, you are a big inspiration for all kit builders. Thanks for the tip regarding removing components. On double layer boards it could be a real struggle. 73, Bas

UTØUM said...

Hi, John
many thanks for your review of my tuner.I hope that it will be very useful for all mcHF's users
I want to tell а few words regarding memory.If we want to organize memory for the tuner, we must know the TX frequency. TX freq we can received from CPU of the transceiver (as in KX2 or KX3),or we can build some kind of frequency counter in the tuner.

In the first case we can't receive any data from mcHF's CPU without modifying mcHF's FW. In the second case we do not have room at the tuner's PCB to organize a frequency counter.The tuner does not have own SWR meter for the same reason.
This is why there is no any type of memory in this ATU.

Regarding drop RF power during tuning.We can't communicate with mcHF' CPU without of modification of mcHF's FW, and mcHF radio does not have any ALC circuits. But RD16 is very strong MOSFET and short increase of SWR can't damage it. You can setup TUNE level at mcHF's menu to 5W. I do not recommend setting up TUNE level to less than 5W, because there would be low level of voltages at SWR pins and tuning will be less accurate.

73! Alexey UTØUM

John AE5X said...

"You can setup TUNE level at mcHF's menu to 5W."

Absolutely right, Alexey - this will guard against accidentally tuning at full output power. Thanks for the reminder...I've modified the text to reflect this.

73 - John

Petr, OK1RP said...

John, many thanks for your excellent review including the pic installation tutorial. It is very helpful and motivating for other builder. Most important are your comments about the memories and SWR readings... In all cases I hope that you will enjoy your upgraded mcHF radio! Good luck, Petr.

John AE5X said...

Hello Petr - thanks for the comment. I'm hoping to use the mcHF this weekend in a portable operation but also have to work about 20 hours this weekend so may have to postpone. Regardless, it's good to know the rig is there waiting and ready.

73 - John

John AE5X said...

There are much better inspirations than me, by a long shot. It was fun though and gave me an excuse to open up the mcHF.

73 - John

ve3axl said...

Another consideration is the use of the ss picks that the dentists use. They get tossed from time to time and a simple request can supply you with probes and drill bits for the Drexel.

John AE5X said...

Good idea, I hadn't thought of that. Toothpicks are a one- or two-use item - after that, they're too charred and weakened to use again.

73 - John

Dirk said...

thank you for your fantastic report!
My 0.6.3 boards were sent with BNC antenna plug instead of the SMA plug. The soldering points for the neon lamp are UNDER the BNC plug's body, so it could be best mounted lieing on the other side of the RF board.
With the ATU, the neon lamp can stay there and the 2 wires of the ATU antenna input could be soldered to the BNC plug wires.

John AE5X said...

Hi Dirk - I'm glad they've reverted back to the BNC jack instead of the SMA. I have to rememeber to bring an adapter every time I take my mcHF somewhere to operate.


nu7b said...

I'm glad I looked at this blog post John - I see that you have soldered the fastening nuts to the heatsinked devices. That saves some rework which would eventually damage the board. I'm getting ready to put my 0.6.3 board back in the case after the following mods:

- 5V 500Khz synchronous switching converter replacing the 7805 regulator (Polulu D24V10F5)
- PIN diode bias circuit mod documented by Ron NU6F
- Replacement of power/SWR detector diodes with 1N5711 Schottky diodes

With respect to the swr accuracy - the diodes on the board's BOM are general-purpose Schottky rectifier diodes with a reverse bias capacitance exceeding 60pF. The 5711 is a small-signal diode with 2pF capacitance. I think the diode C is causing some of the SWR inaccuracy, and I notice after replacement that I'm seeing 1:1 readings now into a dummy load - although I haven't yet tried all bands. Still need to recalibrate when I get the case buttoned up.