Tuesday 17 May 2022

More on masthead preamps.

Can it really be three years since I last opined on masthead preamps? I rarely take so long to ramble on again about things.

http://gm4fvm.blogspot.com/2019/03/more-tropo-masthead-pre-amps-and.html

This time I wish to drone on about how I use the sequencers I have which were made by Down East Microwave Inc (DEMI) in Florida USA. It seems like time for some practical examples.

I have ended up with three of these sequencers. After failing to make a decent job of wiring them up using trailing leads, I built the first one I built into an ABS box with sockets on the outside. The second one had the same box but the sequencer is outside fed by a multicore cable. The third and hopefully final one is built into a much smaller box.

The project to interface the sequencer to the cables is simple. A very low level of technical expertise is needed which can be demonstrated by the fact that I could do it. 

If anyone else wants to try this, the exact wiring they need may well be different from mine, depending on whether they need to run preamps, fans, transverters, or whatever. The DEMI instructions explain which pin does what and I shall not delve into great detail about that.

The DEMI sequencers are available as kit for $60

https://www.downeastmicrowave.com/product-p/ltrsck.htm

or built for $80

https://www.downeastmicrowave.com/product-p/ltrs.htm

As I live in the Scotland there is the International Postage to add, plus possibly also tax at this end if they charge it.

The W6PQL sequencers are good too, and they are installed in my Gemini linears. He sells a rather similar four event sequencer kit which seems to be larger and does not come with a box - but it is only $37.50. You can also buy sequencers that just plug in ready for use. SSB-Electronic make one suitable for 6m to 70cms with built in bias-tee for over £200. I need three, and one of those is for 23cms, so cost comes into this as well as frequency range.

Having chosen the DEMI version, I completed the job on the cheap. These hook-up arrangements I made were built out of my bits and pieces boxes and seem to work fine, and all I needed to buy each time was the basic unit from DEMI. The only exception was the multipin plug which I changed from the one supplied by DEMI.

The DEMI sequencer is not what I would call a stand alone unit. It is a circuit board in a case which is terminated in a 15 pin plug of the type once used on computer monitors. You need to interface to it. I started off just trying to wire directly into the plug, got into a pickle, and then decided to box one up. After that I built the other two as break-out boxes allowing for sockets, switches and fuses on the box which supplies the DEMI unit down a multi-core screened cable.

A DEM sequencer showing stock configuration.

The DEMI sequencer is very easy to wire up for a multiple range of different tasks. It produces four outcomes, all timed to occur sequentially after pressing the PTT and in the reverse sequence on releasing the PTT. Note that, unlike some sequencers which connect into the microphone lead, the DEM ones work on the PTT output from the radio and thus will work with data modes.

There are five options as to what happens at the four stages. These are 

a) Power on during receive (for a preamp) = DC voltage at max 2 amps.

b) Power on during transmit (to power a relay or fans) = DC voltage at 2 max amps

c) Grounding another voltage (to activate a linear amplifier PTT connection)

d) Power on during receive = DC voltage at max 50mA

e) Power on during transmit = 0.7V DC at max 50mA.

You can set the wiring to do these in various (though not all) combinations for the four sequential actions. The ready-made version comes with a standard wiring layout which will suit most situations and you can change that later if you wish. Certainly in my case I do not need all these options and the standard wiring on the built unit is fine for me.

The DC voltage can be between 9 and 17V.

In order, the standard sequence on grounding the PTT line in (when you press the mic button to transmit and your radio grounds its PTT out line) is:-

1) Turn off the voltage to pin 1 (turn off the preamp), then after a pause

2) Turn on the voltage to pin 2 (turn on fans, TX power circuits, etc), then after a pause

3) Ground pin 3 (activate a linear amplifier PTT), then after a pause

4) Ground pin 4 (activate a linear amplifier PTT)

The sequence is reversed when the PTT line in is not grounded, so when you let go of your mic button the steps occur in the order 4 -3 -2 - 1 cutting off power to pin 2 and turning back on the preamp via pin 1.

Pins 3 and 4 could be used to switch a transverter and a linear amplifier in the correct sequence.

The timing gap between the steps is fixed and not defined, but I would guess it is several hundred milliseconds. It is certainly a noticeable between PTT activation and each step after the first one. As the unit is entirely solid state there is no relay delay to add to the sequencing delay, though there is in my linears. It is important therefore to think about the total delay if you are using FM or data modes where hot switching could be a problem. All my radios have adjustable TX delays to account for this.

When it came to the practical installation, the "pin" activations referred to above happen via a DB-15 connector. To begin with I simply wired the unit up in whatever way suited using a DB-15 plug which is supplied with the kit. This requires soldering of a three layer 15 pin computer plug which I never find a very happy task. Then I needed to terminate the resulting wire output with whatever plugs and cables were needed.

This rather ramshackle arrangement ended with me losing a sequencer for some unexplained reason. I suspect that I must have accidentally shorted the coax while working outside on an antenna and this caused something to fry in the sequencer (You should have turned it off first, Jim, so fit a switch and an LED). I had fused the unit on the supply side, but crucially at 3 amps. 3 amps on the input is the recommended fuse as stated in the rather vague instructions. As my preamps draw about 300mA I now fuse the output at 500mA.

It became clear to me that this trailing wire system is not likely to protect the units sufficiently in the environment that I run. Thus I took a sequencer and built it into an ABS box. I decided that it needed (a) fuses on both the input and output, (b) reverse polarity protection, (c) properly fixed sockets so that PTT lines could by fitted (d) a switch so that I can easily turn off the voltage in the coax while working on the antennas and (e) some sort of LED to show it was working. The sequencer has a red/green LED to show it is working but as I was burying it in a box that would hardly help unless I drilled viewing holes in the box which seemed a bit pointless.

The point of the box is really to provide mechanical stability. I found that flexing of the wires in and out of the BD-15 plug would pull the solder joints apart, causing unreliability. The resulting unit mounted in an ABS box looks a bit over-engineered but it has survived for three years without incident apart from the odd blown fuse. It need not be a thing of beauty but it will rarely be seen behind the shelving.

The first boxed DEM sequencer at GM4FVM (the lid is on back to front)

The sockets on the back are phono 1 (PTT in) phono 2 (linear PTT out), 2.1mm DC in, on the side is a phono for preamp out and on/off switch, and on the near side are two fuse holders, on for input (1000mA) and one for output (500mA). The front panel has two LEDs for transmit and receive. On the top you can just discern the trace left by a tea cup, showing that this item is multi-purpose.

I needed sequencing because I use an IC-7100 for 432MHz, and that radio cannot supply the voltage for the preamp. I then acquired a splendid linear amplifier for 1296MHz. This had previously been used by Sid, G8SFA, and I drive it as a second stage with my existing 1296 linear fed by the IC-9700.

Whilst the 9700 can supply the voltage for the preamp, I could not arrange the two 23cms linears to switch simultaneously (one has a sprung antenna relay, the other a latching one). It occurred to me that the DEMI sequencer could switch two linears via pin 3 and pin 4, and that all the output pins are isolated. I had a spare sequencer left since I had used one for 2m with a transverter, so I put it to use. I needed to rewire it as I had never used pin 4 before but it all worked fine.

This second installation actually used the first sequencer I bought and this was the one which had been initially wired up with trailing leads which was an unreliable method to use. At first I thought of putting this one in a box too. That seemed crazy so I decided to use the box as a break-out box, leaving the smaller sequencer unit itself at the front of the bench. The larger break-out box will lurk at the back out of sight. This one has the same plugs, fuses and on/off switch as the first one, with an extra socket for "linear 2". There is still an LED, but just to show that the preamp voltage is reaching the break-out box. I did wonder if this LED was necessary, but I managed to blow the output fuse at an early stage and the LED proved where the problem lay.

Now using the two timed grounded outputs on 1296 means that there is a slight delay between the two linears switching, but this does not seem to matter at all. The advantage is that they are isolated from each other. The preamp is powered directly from the 9700 so no need to use that socket for now, but it is still available should I move it somewhere else. I also used a DB-15 plug with screw terminals just in case I wanted to alter the pin wiring later too.

Screw terminal DB-15 plug saves soldering 3 rows of tiny pins.


There was a slight delay in building it when it turned out that the DB-15 plug with screw terminals, which I had been safeguarding for this purpose for years, turned out to be male and I needed a female one. D'oh. They come from China and take weeks to arrive. Luckily I could buy a cheap "MINI GENDER CHANGER" to tide me over. Because I had not discussed my plan in this direction with her, Mrs FVM was a bit surprised to see a gender changer arrive in the post for me.

My MINI GENDER CHANGER & sequencer - what a relief.

I might point out that a lot of this effort would have been saved if DEMI had used similar pin connections to those on standard SVGA display leads. These leads are terminated in DB-15 plugs. Long ago I thought I could just cut an SVGA lead in two and use both ends as two sequencer leads. Sadly standard SVGA leads are not "all pin wired", and several pins you could need for the sequencer are combined as ground connections. Ironically, I ran out of the 9 core screened cable for my next sequencer project and I had to use an SVGA cable instead - I had to cut the DB-15 plugs off and throw them both away and then connect one end to a new DB-15 plug.

Next, for another project I needed a third sequencer. Once again I have bought a DEMI one. I have finally decided to use a smaller box. And, oddly, I am using the same way of getting the cable from the sequencer into the box - an ugly gland. For the second sequencer the original plan was to use an 8 pin DIN plug and socket. Then it occurred to me that I was never likely to disconnect it and anyway there is a plug and socket at the sequencer end. So I delved into the "gardening electrics" box hoping for a grommet and found instead a waterproof gland, total cost zero. So if the tide ever rises 60 metres above mean sea level at least that aspect of the GM4FVM set-up will be safe. Mind you the water will just get in down the centre of the phono sockets instead, but more gradually, allowing me time to bale it out again with a spoon.

Break-out box for third sequencer under construction

This new smaller box features the same over-sized (but cost free) gland for wire access, on/off switch, LED plus output and input fuses. The five sockets are on the rear panel. At last this beginning to look a bit neater.

I suspect that the DEMI sequencer is based on the circuit used by them on their transverters. As such I guess that most of the ones they supply get built into other equipment and not too many end up stand-alone. My use of them initially standing beside the rigs with trailing cables simply connected into the plug on the back was pushing things. I think they really need a break-out box if used in that way, even if only to protect the wiring and tidy things up.

My preamps all have RF VOX switching, which I regard as a back-up safety system. I know that some people rely on using this to switch a preamp or transverter rather than installing a sequencer. When I have tried it I can see the SWR spike in the rig when going to transmit. There is a risk here of either this damaging the output stage of the radio, or of hot switching damaging the preamp. Given the cost of preamps and radios compared to the cost of sequencers, I know which solution I would choose.

Box number 3 being tested.

The first basic one has been in continuous use on the 70cms radio for 3 years. The second one has seen a couple of stretches of use, and I am happy now that it is tidied up and working on 23cms. The third one ... well you will just have to wait and see what that project is.

Finally, with the last sequencer came a strange note from DEMI stating that the latest release has a wrong package style for Q2 installed causing incorrect operation of sequenced step 2. It did not explain whether this applies to my one, only that kits and some assembled units dating back to Dec 2020 are affected. It states that "the assembled units will pass out test fixture testings to verify switching but the units will not operate correctly with a current drain above 10mA". It then goes on to show a modification using a leaded 220 ohm resistor to replace SMD resistor R2 in a different place. This looks like an easy modification to make.

I was not exactly sure if that meant the unit I had was affected. If I did the modification would that stop my unit working if it was not affected? Anyway, I decided to test it with the masthead preamp which draws about 300mA and it worked fine. For the moment I will leave it at that.

73

Jim

GM4FVM

Wednesday 4 May 2022

When is a 50MHz opening really a 50MHz non-opening?

I am a great one for watching the Internet for opportunities to work DX.

Many sources of intelligence are used at GM4FVM, Solarham, PSK Reporter and often also the excellent DX Maps.

DX Maps is very useful on three counts. It has contact maps so that I know where to point my antennas. Then secondly the MUF Sp-E map is derived from that information and it shows the predicted maximum usable frequency for locator squares over Europe. This means that I can see Es developing and, perhaps more importantly, fading. And finally it shows lists of QSOs reported and also allows me to post formatted reports myself.

Here is a map showing all the activity as reported to DX Maps

DX Maps showing all reported 50MHz activity on 2 May 2022

Despite this there was no sign at GM4FVM that the 50MHz band was open.

When I went into the "Select Options" tab on DX Maps and chose "Only DX-Cluster" this was the map which resulted, which was much closer to what I was observing from here.

DX Maps showing "Only DX Cluster" on 2 May 2022

The difference between the two maps is, broadly, that the standard map shows anything reported (which includes automatic uploads), whereas the second one shows entries made by amateurs manually posting things to the Cluster. Not all of the manual postings to the cluster are for actual QSOs, some report simply stations heard, usually unusual ones. However, most of them are QSOs. On the other hand, the upper graph contains dozens and dozens of automatically generated reports many of which are simply single decodes.

The thing about this time of year on 50MHz is that there are a lot of stations calling CQ but not many having contacts. So these CQs get picked up and reported automatically by stations normally on receive. There is nothing wrong with this, but it tends to clog up the reporting system. Just decoding a single FT8 message (most are FT8, but not all) does not make a QSO. Not only is the E layer very unstable at this time of year, there is also the Eta Aquariids meteor shower which often generates these single decodes, plus of course random meteor pings.

Just selecting "Only DX-Cluster" often makes a dramatic difference. The effect can be seen on the "List" section of DX Maps which shows the postings in tabular form. Here is the latest one and as usual with images here you can click on it to enlarge the text.

DX Maps List on 4 May 2022.

When I clicked "Only DX-Cluster" every one of these red automatic entries disappeared to be replaced by a list of manual postings in black type. Basically, everything recorded had been a single decode and in one case the same signal decoded on two frequencies at the same time. As far as I could see nobody  was actually having a two way contact. Changing to manual postings suggested that, rather than several contacts every minute all over the Continent, there had not been a QSQ recorded for more than twenty minutes and the ones that were had been largely concentrated in a single area in Germany.

DX Maps is not the only site affected by this effect - PSK Reporter is too. In fact PSK Reporter is affected even more because you cannot switch it to "Only DX-Cluster". In that sense DX Maps adds a very useful ability to screen out almost all these one-off reports. Of course there will always be the odd one-off report, often due to random propagation enhancements, possibly caused by cosmic ray ionisation, orbiting objects, meteors, aircraft scatter etc. They can be quite fun to see on the higher VHF bands, but they never seem to last long enough to work someone. If only. I still reply to a few just "to be sure".

The very useful MUF Sp-E chart on DX-Maps seems to always include all the reports and you cannot screen out the "red" ones. It works very well when there is only really Sporadic E and not the random meteor scatter pings we are getting at the moment.

It is very handy to see the one off ones too, but it is easy to get carried away and think that there is an opening when there isn't.

For the record there have been a few short Es openings on 50MHz at GM4FVM but nothing on 70MHz yet. 

As for 144MHz Sporadic E, well that will have to wait until ... (who knows?).

73

Jim

GM4FVM