Merits of The Wireless Set No. 19?

[Argus25]

Split from this thread:-

https://www.vintage-radio.net/forum/...d.php?t=155632

What a great set the 19, these remind me so much of the ZC1 (which I have, but not the 19 yet). I think the 19 has exactly the same pocket watch holder as the ZCI. The front part of it is a ring that unscrews with fairly fine threads around the circumference, sometimes they can lock up and be tight to unscrew.

I had to hunt around for a while to find a WW2 military pocket watch that would fit in the holder. Many pocket watches of the era, the geometry of the winding shaft is wider and they were such that they would not fit in the shell without modifying it (which I didn't want to do). I've attached a photo, sorry poor quality image. In the end I collected about 5 pocket watches and got interested in those too. It is also fun to restore a pocket watch, its not as difficult as a standard wristwatch, as everything is bigger making it much easier to handle the gears and balance wheel etc.

[trh01uk]

Originally Posted by Argus25

What a great set the 19........

Was it? Its a set very popular with collectors because a) it looks the part (being totally unlike anything to follow later,) and b) its widely available.

Its further claim to fame is that it was widely deployed by the British Army, in a huge variety of vehicles, and over a very long time period from the early 40s right through to the 1960s. Even longer if you count the new ones I was still pulling out of the original packing as an army cadet in 1973! And then it was made in various countries, UK, USA, Canada to name the obvious ones.

As a piece of design work it reflects the tale that it was "designed in 6 weeks", by Bill Pannell, a young postgrad at Pye. So its a good indicator of what industry could do back then without any time for serious innovation. That was the Mk.I of course, and considerably more work was required to produce the Mk.II which could actually be deployed and used in field.

It failed in several ways. The VHF B-set was first informally abandoned by users as "useless", and then formally abandoned by the Army in the 1950s, when they mounted to programme to rip out the useless electronics, which could readily consume power for no benefit. The multi-way connectors are flimsy and unreliable.

If you break one of those multi-way connectors on the set (easily done) you then have the joyous, multi-hour job of changing it. Which points to another shortcoming: servicing for anything beyond the trivial (e.g. changing valves) is something of a nightmare, with components laid over each other under the chassis.

As for performance, all one can say is that it was better than what had come before, so it had much better frequency coverage (2 - 8Mc/s) compared to the WS11 (4.5 - 7.1Mc/s). The ability to go down to 2Mc/s would give far better ground wave performance, which was critical between vehicles with shortish vertical rod aerials.

Its worth noting that the German kit at the time was vastly superior, and it would take us some 20 years more development for us to catch up. The continuation of the WS19 for so long also reflects the social history of the UK - the country was bankrupt, and the army had to "make do" with barely good enough kit for several decades after WWII.

So an assessment of the WS19 rather depends on your definition of "good"...

Richard

[ex seismic]

A bit harsh there Richard. Like a lot of wartime equipment it was designed to do a job, in a hurry. A bit like the R1155 I suspect the service life was not expected to be long enough for serviceability to be a major design factor. Having said that I'm sure a better component layout could have been achieved if only to make assembly easier. The WS19 is an obvious ancestor of Larkspur and Clansman in that is is part of a system, not just a radio. I regard it as being the first genuine dual band radio, and with memories even!

Gordon

[G6Tanuki]

Originally Posted by Argus25

Many pocket watches of the era, the geometry of the winding shaft is wider and they were such that they would not fit in the shell without modifying it (which I didn't want to do). .

Yes it was quite common to find used 19/52-sets etc with a bit of bakelite cracked out of the watch-holder body to allow the fitment of a non-standard pocketwatch.

[The radios that came with a watch-holder - I generally removed it and used the space revealed for something more-practical, like an extra switch, pot or socket. At least the modification could be removed and the watch-holder refitted to cover the hole if in future times 'original appearance' ever became an issue] .

[trh01uk]

Originally Posted by ex seismic

A bit harsh there Richard. Like a lot of wartime equipment it was designed to do a job, in a hurry. A bit like the R1155 I suspect the service life was not expected to be long enough for serviceability to be a major design factor. Having said that I'm sure a better component layout could have been achieved if only to make assembly easier. The WS19 is an obvious ancestor of Larkspur and Clansman in that is is part of a system, not just a radio. I regard it as being the first genuine dual band radio, and with memories even!

Gordon

Harsh? I don't think so, because I am trying to make an objective assessment of the radio, regardless of the circumstances it was born in.

Look at it another way. To produce any radio that complex, with the tools available then, in 6 weeks is an astonishing achievement. Particularly for a company previously turning out domestic receivers. I bet the team had to work 18 hour days, 7 days a week just to produce a few samples that actually worked.

And "actually working" - when the trials of sets came along was what actually counted. The SEE sample sets - three in all - had faults in two of the sets, which made it impossible to fully evaluate them. By contrast Pye produced 10 - 15 samples, and they were good enough to allow a full trial. At the same time, the Pye set did not meet the WD specs. It only covered 2.5 to 6.25Mc/s against a spec of 2 - 8Mc/s.

The need to introduce bandswitching to achieve the required frequency coverage was a serious drawback - see the long note in WftW about it. Reliability is cited as one of those drawbacks.

Regarding serviceability, my bet is they never even thought about it once! Just getting 10 working samples in 6 weeks is enough of a challenge. But don't assume serviceability was not important - you only have to read the extremely extensive EMERs, with even instructions how to rewind the iron-cored components - to see that repair of just every part was required. In a war, the idea of swopping radios in and out of the field becomes pretty unlikely - you have to fix what you've got!

There is nothing novel about the WS19. The single frequency tuning system came in with the WS11 years earlier. You mention it being part of a system - I presume you mean the vehicle harness. That was doubtless part of the design requirement, but it has little to do with the WS19 and mostly to do with the vehicles, how noisy they were, and the need to communicate between vehicles and ground stations from anywhere in the vehicle (not just the radio op).

The "memories" were already featured in earlier sets. WS9 I think was an example.

Larkspur was not based on the WS19, but on the WS42, which was the first fully hermetically sealed 2-way radio in UK, with a diecast chassis. That set was designed at SRDE (formerly SEE) during WWII. My suspicion (lack of historical records preclude proof) is that it was heavily based on German radio designs, which by then had been captured and closely examined, and which had incorporated these principles from pre-WWII. The Germans were some decades ahead of us, and had the luxury of time to design their equipment. We were caught "with our pants down", and thus had to go with "whatever we could knock up in the circumstances".

Richard

[Argus25]

Originally Posted by trh01uk

Was it? Its a set very popular with collectors because a) it looks the part (being totally unlike anything to follow later,) and b) its widely available.

Excuse the brevity , I should have said a "great looking set" as I have not looked in one or repaired one before, so any shortcomings I was not aware of.

It so reminds me (cosmetically at least) of the ZC1, which I am a devoted & loyal follower of and is one of the most astonishingly well made sets I have ever seen with a manufacturing quality that is excellent (and I am a very harsh judge of construction quality too and not that easily impressed). So it is disappointing to hear that the 19 set has the issues you mentioned. But hopefully it has some redeeming features that would add a fun element to it for a restoration project.

[thx1138]

Hi again,

I added a few more pics to my folder in link above. I got hold of the No9 key and also took plunge and started the repair of the headphones. Not a fun job, both speaker elements were open circuit , and (bonus) the plug on end of lead had 2 broken wires also which had to be resoldered, no doubt due to age. My biggest battle wasnt the less than nice resoldering emergency repair ,yes it looks bad, but the wire was next to impossible to see.. However its all done and both come up as 48 ohms each and are testing fine now.

The headphone earpads : Managed to slip both off without them falling apart but Id love to replace them. Theyre really degraded.
Microphone seems ok and so far so good... The control box is fine too, so next is the psu and radio caps to do. As there are a number of options as to What to use, can anyone suggest the best longterm type or any recommendation before i go shopping for caps.. Also thanks again for help & info. Re radio itself, rough and ready Id say. I had a 62 set,and it was a lot less hassle to work with and restore.. I guess its age and progress..

[BK0scillation]

Hello,some years ago I purchased some 0.1 and 0.01 tubular capacitors
they fitted the clips perfectly on the 19s sets they were rated at 1kv
These were from RS I am sure that they do not have theses any more
I will check around probably buy a quantity, the wound components
are a problem as the input and output transformers have very fine wire,
that is almost impossible to wind (I have no doubt that some one has been successful)
The are great set fun to restore and to use I have a few and the best one has had a really rough life before restoration but has given really sterling service,
Good luck with your restoration

[dave walsh]

I suppose the pocket or regular watch is a good metaphor for these sets ie somewhat complex and difficult wiring wise and in other ways. You could buy a valve-less chassis for 3/- at one time and many of us did because we could afford them but I never tried to get one running-interesting to dismantle-especially the dials!

They did seem clunky and hard work but the regenerative B set [a valve and a coil] was modified by some as a VHF device. Others got the main set operational and enjoyed the experience [there was a complete restoration article series in Practical Wireless]. I wouldn't have wanted one in my tank though! As trh01 said, they were produced in times of urgency and not really expected to survive [in operation] to need servicing. I was very surprised when they became such prized items over the years but there may be lots of reasons for that-maybe including the complexity and getting all the accessories etc together or just a relative shortage of complete sets? They did look intriguingly technical on the page. Who knows how these things come about really. It wasn't anticipated that Round Ekco's would be venerated when they were first manufactured, they were just a basic set for the home!

From the point of view of an enthusiast hoping to get a decent comms receiver, cheaply, out of the post war Government Surplus flood, the R1155 is a useful comparator to the 19 Set. Also a rush job, floating negative chassis, tight wiring and not designed for service [given the bomber fatality rates]. It had a proper half moon dial though, several switched bands and the right look along with the desired communications receiver features, with space to add on or "improve" if needed when the DF section came out [another similarity if not a B set]. I didn't get one for a very long time, though, as 1155's were still not that cheap Even the majestic AR88 has a built in potentially disastrous PSU weakness!

Dave W

[G6Tanuki]

Must admit, I've never really understood the mystique and desirability of the WS19; to my mind it has a number of significant conceptual flaws:

1] Horrible tuning-mechanism. Try setting one to 7.045MHz and getting it sufficiently-right that when the rest of the net comes on-frequency your WS19 is sufficiently close to the required frequency that it's within the RX passband. You might just get it right in a static situation, but not when you're bouncing around in a tank!

2] Takes a hell of a lot of DC power but produces very little RF. I guess this might have been deliberate in that it means the 807 isn't likely to die from flashovers or anything if the antenna is massively mistuned.

3] The frequency-coverage doesn't really make it easy to get good antenna radiation-efficiency: even at 8MHz the standard 16-foot whip is only 1/8-wave, at the LF end it's very definitely a 'short' antenna, and base-loaded too, so its radiation efficiency must be gruesome!

The US SCR-694/BC-1306 was a lot more-compact but still had the issue of less-than-ideal frequency coverage for mobile use; their BC-659 was a much more-sensible vehicle radio, with 30-odd Watts of RF output, crystal-control, FM, and operating on frequencies (27-38MHz) where an efficient 1/4-wave whip antenna is perfectly practical. Think of it as the first "CB" radio!

[Argus25]

Originally Posted by G6Tanuki

2] Takes a hell of a lot of DC power but produces very little RF. I guess this might have been deliberate in that it means the 807 isn't likely to die from flashovers or anything if the antenna is massively mistuned.

3] The frequency-coverage doesn't really make it easy to get good antenna radiation-efficiency: even at 8MHz the standard 16-foot whip is only 1/8-wave, at the LF end it's very definitely a 'short' antenna, and base-loaded too, so its radiation efficiency must be gruesome!

Interesting points, the frequency coverage is similar to the ZC1 too, but it uses a 6V6 (not 807) and ideally matched out of that I can get exactly 3.26w on 80m and 2w on 40m, so I would think the 807 could do better than that ?

I have seen just as many photos of these sorts of radios set up on the ground as base stations as stuck in a jeep and I do wonder if there might be some advantages of the lower frequencies compared to the 27 to 38MHz area (although the antennas are more awkward) with less absorption by plants/trees & bush. CB radio signals tend to get very heavily attenuated by trees and bush.

Although I do think it was a bit awkward that you had to lug around lead acid batteries (at least for the ZC1). I assume that the WS19 would have run off 12V too.

Still, I think the notion of energy efficiency barely applies to these sorts of radios, as long as they could furnish some communications across distances that were further than bullets/shells could travel.

[G6Tanuki]

According to the Pye Museum, the WS19 had a "Transmitter RF output: CW 3-5W or greater, AM 1.5-2.5 Watts or greater (Note there are wide variations in RF output between sets)"

Hardly impressive, I think you'll agree: even a "high achiever" radio that managed double those figures wouldn't exactly be taxing the old 807.

I think the UK equivalent of the ZC1 would probably be the WS22 (which had the same frequency-range as the WS19 and, produced a couple of Watts of RF and had a built-in ATU), or later on in WWII the WS62 (which covered 1.5 to 12MHz and gave a Watt or so of output).

[trh01uk]

Originally Posted by Argus25

Interesting points, the frequency coverage is similar to the ZC1 too, but it uses a 6V6 (not 807) and ideally matched out of that I can get exactly 3.26w on 80m and 2w on 40m, so I would think the 807 could do better than that ?

I have seen just as many photos of these sorts of radios set up on the ground as base stations as stuck in a jeep and I do wonder if there might be some advantages of the lower frequencies compared to the 27 to 38MHz area (although the antennas are more awkward) with less absorption by plants/trees & bush. CB radio signals tend to get very heavily attenuated by trees and bush.

Of course there is an advantage from the lower frequencies. We are talking about ground wave propagation here - not a mode that's popular any more. The lower the frequency the better the signal propagates - in general. Some of the really early tank sets were operating down in the 100s of kHz region, way below anything the WS19 could produce.

The 2-8Mc/s band was where most radio comms took place during WWII - at least amongst the Allies. As a result competition for channels was fierce, and at night time skywave interference down at the bottom of the band could be severe. That's why the Germans had the sense to move much of the short range comms to 30Mc/s and above. They had some infantry sets around 100Mc/s - we tried such VHF sets and failed.

In the UK the RAF had much better success. During the Battle of Britain they were at least trialling the TR1133 VHF set in planes, talking back to a giant T1131 transmitter with accompanying R1132 receiver. Most of them were still using TR9 HF sets though with all the problems of interference.

There was a strong belief amongst the military that "VHF cannot go around corners". So they thought that on a typical battlefield VHF would be useless. They were completely wrong of course - but that's what you get when you have a body of people dominated by people who probably did Philosophy, Politics and Economics at uni, rather than engineers who might think more appropriately. It took the Americans with the BC-1000 backpack, operating 40 - 48Mc/s to arrive on D-Day and demonstrate the vast superiority of that set of the British WS18 (also operating at HF).

We might note that the WS19 actually provided about 3 - 5 watts on CW, which can generally do 10 miles or so down around 2 - 3Mc/s (actual range depends on many factors including ground conductivity). But on voice, the output is cut drastically, because the 807 is biased back, so it provides 5W on voice peaks. But the carrier itself - the usual measure of AM output - is only a quarter of that (theoretically) - so an output of 1.2W is fairly typical for the carrier.

Richard

[trh01uk]

Originally Posted by G6Tanuki

According to the Pye Museum, the WS19 had a "Transmitter RF output: CW 3-5W or greater, AM 1.5-2.5 Watts or greater (Note there are wide variations in RF output between sets)"

Hardly impressive, I think you'll agree: even a "high achiever" radio that managed double those figures wouldn't exactly be taxing the old 807.

I think the UK equivalent of the ZC1 would probably be the WS22 (which had the same frequency-range as the WS19 and, produced a couple of Watts of RF and had a built-in ATU), or later on in WWII the WS62 (which covered 1.5 to 12MHz and gave a Watt or so of output).

Yes, I agree with you that the output from these sets is distinctly unimpressive. If we take the upper limit of output - a mere 5 watts on CW - then that is certainly poor. The DC input to the PA - with 500V on the anode, and current of 35mA (that's only from memory though!) is 17.5 watts. Assumming a very inefficient stage, say 50%, we would still expect something near to 9 watts output.

Strangely, when you examine an actual EMER you get a somewhat different picture. Take EMER FZ254/3 for instance (which happens to be the Canadian WS19 but I don't think that matters), it has (clause 158 and Fig.86 for test circuit) a procedure where the output is taken via the variometer into a dummy load consisting of a variable cap (to cancel the variometer reaction I guess), a 24 ohm resistor and an RF ammeter.

The spec on the minimum RF current is generally 0.63 amps at most frequencies from 3.5Mc/s upwards (slightly less at 2.5Mc/s).

When you work out that current into 24 ohms you get 9.5 watts!! That looks much more respectable and in line with my calculation above.

So what's going on then with this figure of 3 - 5 watts? Its certainly what I have normally seen when a WS19 is plugged into a typical 50 ohm power meter and run on CW.

Is 50 ohms a severe mismatch - and thus failing to properly load the PA? (There's no control of loading of course with the WS19 PA circuit).


Is there not enough PA drive, thus failing to push the 807 into class C? And of course then dropping the efficiency of the PA severely?

Perhaps someone with the time, and appropriate test kit could have a look at this!


Richard

[paulsherwin]

They certainly soldiered on. The BAOR was still using them operationally in the late 60s as HF transceivers, though in small numbers and not in vehicles. The cadets used them well into the 70s.

[Bazz4CQJ]

I remember having a Saturday job with "Mac of Morley" in the late 60's and he bought lorry loads of them. They were all sold at a standard price of 39/6 (that format looks wrong, 39 shillings and 6 pence) and some of them were in immaculate condition. Personally, I was never drawn to them and never owned one.

B

[G6Tanuki]

Originally Posted by paulsherwin

They certainly soldiered on. The BAOR was still using them operationally in the late 60s as HF transceivers, though in small numbers and not in vehicles. The cadets used them well into the 70s.

I struggled with a WS19 on the Army Cadet 60-Metre nets in the 70s; I was rather glad when COD Donnington released some C12s, even though it did involve some interesting 'negotiations' with the chemistry-master to get a couple of gallons of H2SO4 for the pair of metal-cased 110AH batteries needed to power it (C12 was 24V).

The C12 was great! Proper high-level modulation and a nice ATU. I worked all over the UK during lunch-breaks, that being the only time when AF breakthrough into the intercom between the headmaster's study and his secretary's office was not going to be a problem.

I had a 22 set when a kid, interesting design, discuss?

[VT FUSE]

My Father was a member of a Royal Signals Regiment based in Palestine during the Mandate period and British withdrawal/policing operations etc in 1947-8 (no politics etc inferred here,please) and when he was alive he occasionally spoke of the radio nets that operated using WS19/52 sets etc and commented on the long distances covered via Ionospheric propagation,certainly in the hundreds of miles and once spoke of hearing post ww2 British amateurs being heard and his firm belief that two way qso's may have been possible, but of course,military to civilian amateur radio traffic was disallowed and enforced under threat of severe consequences.Also important to consider would be lower levels of qrm in those lesser tech days.

My father bought me a WS19 (Canadian MK3) as a present when I was aged around 14 in 1971 and once instructed in tuning up a 130 foot inverted L Aerial the station would have feasibly enabled me to work inter G on 80M and very likely much of Europe as well (of course,I never attempted to )

I have memories that my father mentioned these radios being powered up for months in the desert and of how they would give early warning of approaching sand storm static qrn and how Aerials would collect strong static charges.

I recall that many WS19 receivers had damping resistors across the if transformers to broaden the if tuning and also to negate slight shifting of the cores during battle in armoured vehicles such as tanks under fire/firing main armament.

Mike

[trh01uk]

Originally Posted by merlinmaxwell

I had a 22 set when a kid, interesting design, discuss?

I suspect we will only be allowed to discuss the merits of the WS22, with reference to the WS19, otherwise we will be off topic.

This actually highlights one of the strengths of the WS19, which many will take for granted. Namely that you can always tune the transmitter and receiver with just one knob. If you get the receiver on channel - then you can be confident the transmitter will be too. Assuming of course you are using single frequency simplex operation!

By comparison the WS22 is not a true transceiver (as defined above). The transmitter has to be "netted" to the receiver by using the "Netting Trimmer" on the front panel (just bottom right of the main tuning control). And if you compare the block diagram of the WS19 and WS22, you will see that the former obtains its RF output by mixing the Rx local oscillator with the BFO - while the WS22 has a separate master oscillator, which is roughly on the same frequency as the receiver, but can easily be some 10s of Kc/s off, unless properly netted.

Given that the two sets look near identical, this seems to be like a serious flaw in the WS22 since the WS19 was the "standard" by which to judge all similar sets of the family. Anyone trained on the WS19 would immediately assume he could drive a WS22. And could then easily forget to do the critical netting of the transmitter because it isn't required on the WS19. The days of carefully designing the "man/machine interface" were far in the future at that time.

I suspect (but I have never seen any written evidence) that the failure of comms at Arnhem during WWII, where WS22s were in use amongst various sets, is likely to be partly due to this "feature" of the set, which were being used by briefly trained operators.

Richard