|
Vintage Television and Video Vintage television and video equipment, programmes, VCRs etc. |
|
Thread Tools |
25th Sep 2017, 6:27 pm | #1 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
The pre-war picture on the screen
The HMV 904 was a low priced set by the standards of 1939, costing 29 gns. or a deposit and 6s 4d a week. It included an all-wave radio.
There's not much room inside. Everything is very much crammed into the cabinet and there are internal heat deflectors. Never a good sign. The picture is small - only 5 inches diagonal if all of it is to be seen behind the circular mask. I suspect the picture was never as good as that available from a 'first generation' (1936-7) '60 guinea' EMI set. As a very early domestic receiver, maybe it's unfair to judge it too harshly, but I will try. Some observations: 1) The photo exposure may not be ideal. 2) The television picture is much smaller than on the first generation sets. 3) The main deficiency is lack of sharpness, though this doesn't seem so objectionable when viewing the actual small picture. The likely reasons: The vision circuitry is a superhet rather than TRF. With standard alignment, the vision bandwidth (Argus25 tells us) is only 1.4MHz. I seem to remember this limitation also applied to the 9-inch second generation '907'. The native image on this original Emiscope 3/1 tube seems to be a little softer than that seen on Argus25's 5FP4. How much this is down to the condition of this tube and how much was original, I don't know. 4) There is field cramping at the extreme top. I have noticed this on other off-air photos on the web of this model. So far, I have checked the values of the field T/B components are correct and added a 0.05uF HV cap across the frame scan coils. I have a copy of the circuit courtesy Brian Cuff that I can post sections of here. Ideas welcome. 5) There is some evidence of deficient LF response. I only spotted this on the photo. It's not obvious looking at the actual television picture. 6) The ion burn is proportionately much larger on these second-generation tubes than on the first. Someone here might like to explain why. Having said that, the viewing experience certainly has a lot of 'vintage' charm. When this set is back in its cabinet, I propose to make up a DVD of late 'thirties Pathe variety acts. This sort of material is very like much of what was seen coming from Alexandra Palace at the time. Expect another off-air picture! Steve |
25th Sep 2017, 6:39 pm | #2 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,998
|
Re: The pre-war picture on the screen
Must admit, I've always wondered about the extent to which the major receiver-suppliers to the emerging TV market would have been able to influence [behind the scenes] the characteristics of the transmitted signal in order to present their particular receivers in a better light ??
[Post-WWII the emergence of 'flywheel sync' to give better 'fringe area' reception highlighted some significant and devastating issues in the consistency of BBC transmissions, and the BBC took some time to yield to/correct the demonstrated technical proof of the inadequacy of their broadcasts] |
26th Sep 2017, 1:15 pm | #3 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: The pre-war picture on the screen
Quote:
The photo exposure is difficult. Camera photo's of a CRT tend to over expose the highlights. I think you did well with the photo. The reasons the screen image is better with my 904, with the 5FP4 is two fold. Firstly I widened the tuning bandwidth of the video stages (different to the manufacturers's recommendations) and the 5FP4 is a superior tube to the 3/1, with a small sharp focus beam, it was designed for viewfinder use. The raster cramping at the top is easy to explain. The problem here is that a trapezoidal wave is not supplied as the grid drive voltage to the vertical output tube, instead its only a sawtooth. Because of this the output tube doesn't cut off adequately at flyback and the top of the raster (early part of the vertical scan) is compressed. It is interesting that the Americans had figured this out at the time in the late 30's, but there was little incorporation of this idea in UK sets. The drive voltage for the vertical output tube driving the yoke, to attain a linear sawtooth current in the field coils, must be a combination of a sawtooth and a square wave. The square wave component being required due to the inductance of the yoke and the sawtooth due to the resistance. Nearly all American sets had a peaking resistor in series with the sawtooth forming capacitor to attain the square wave component and create the trapezoidal drive waveform. If you add a resistor, as shown on page 2 of this article, it will fix the top of the scan: http://worldphaco.com/uploads/CONVER...A_5FP4_CRT.pdf In this set if the LF response is deficient, it could be due to the video RF stage alignment, since it is direct coupled from the anode bend detector then directly to the CRT cathode so the DC component and LF response can't be lost there, unless.... Say the cathode bypass capacitor on the video out tube has gone low value..or C60 that feeds the CRT has gone low value. Also the screen bypass electrolytic on the video out tube could cause poor LF response if low value. So check those three electrolytic caps and if they are ok, then sweep the set from the antenna to the video output on the plate of the video output tube(anode bend detector) to check the bandpass characteristic. Assuming say a similar deflection angle comparing a big & small CRT, the ion burns always look proportionally a bigger % of the faceplate on a small face CRT, and I'm not sure exactly why. Last edited by Argus25; 26th Sep 2017 at 1:21 pm. |
|
26th Sep 2017, 2:36 pm | #4 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: The pre-war picture on the screen
Steve,
Also when I did align my 904, it was a technical challenge to attain a wider bandwidth due to the interactions between the sound & vision stages. So when it was done I measured the projections of the adjustment slugs, in mm, above the round brass nuts, and documented the shape of the sweep. I figured this might help others doing it (attached). If you set the positions like this, it could probably be a good starting point before fine adjustments. It is labelled "IF sweep" but its just a sweep from the antenna input to the video output. Make sure the sweep generator is coupled in by a resistive dummy antenna pad to avoid a spurious response. One odd thing about the brass slugs. If there is any wax on the threads it partially insulates them and there can be intermittent effects due to connection-disconnection of the slug to ground, changing the tuning due to change in capacitance. So I had to clean the threads of all wax and to lock them I used silver conductive paint (from a pen) like the type used for pcb repair. |
26th Sep 2017, 2:47 pm | #5 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
Useful tips Hugo. Thanks!
In the meantime, a new fault has appeared which will need sorting. At first warm-up, the set 'took off' for a short time. This was probably caused by a valve. Next time it happens, I'll try to isolate which stage is causing it by manipulating the top caps. As regards the slightly deficient LF, could this (and the soft focus... vis. it's difficult to see the scanning lines... be caused by the tube being slightly soft? Could this mean the input impedance at the cathode-grid was reduced? In which case, matters might improve with use. Steve |
26th Sep 2017, 2:54 pm | #6 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
And another thing. As regards alignment, I was limited by L28 being completely seized, despite overnight bathing in unlocking fluid from the top + heat + tapping etc. The oscillator originally was actually running at 51MHz(!). I replaced the timing capacitor and set it to exactly 37MHz with a counter. The alignment then proceeded satisfactorily with no problem from sound-on-vision.
|
26th Sep 2017, 7:02 pm | #7 |
Heptode
Join Date: Jun 2006
Location: Ware, Hertfordshire, UK.
Posts: 988
|
Re: The pre-war picture on the screen
Hi Steve,
A couple of points... Electrostatically focused tubes normally produce a very small ion burn. It is a peculearity of the 6/6 that it produces such a large (1/2") ion burn for an ES focused tube. Maybe it is something to do with its length. With magnetically focused tubes it might well be related to the EHT. Smaller tubes typically run at lower EHT, so perhaps this explains why the ion burn is larger on the smaller tubes. It might just be down to the gun construction though. I found that the Russian 18ЛК5Б that I am using in my 7" set, required a lot less focus current than the original 3/2. Regarding the alignment... I found it very hard, in fact impossible to do properly with the spot frequency method due to the heavy damping on the vision IF coils. I will be doing it again in the same way that Argus25 did by using a sweep generator. In fact I have recently purchased an arbitry signal generator for that specific purpose. Cheers Andy
__________________
www.youtube.com/user/andyvalve100 |
27th Sep 2017, 8:48 am | #8 | |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: The pre-war picture on the screen
Quote:
It might not actually be a fault or a valve problem. I think I recall in my set, while I was adjusting it, that some combinations of the tuning slugs were unstable and the amplifiers burst into oscillation. I think the reason behind this is that the shielding from input to output of a stage is not 100% perfect, so when the output and input resonant circuits are at similar frequencies they can exchange enough energy with each other to sustain an oscillation, despite the isolating effect of the screen grids. I also found that using the setup procedure for the tuning in the manual, following it exactly, gave a very disappointing result when I checked with the sweep generator, with a bandwidth of only about 1 to 1.4 MHz and not making best use of the tuning coils/stages that were there. They must have been much more interested in gain than bandwidth/picture quality. I found that even with the wider tuning, it can still produce a reasonable image with peak RF input voltages of 500uV to 1mV, but I tend to feed it with 3mV peak signal from my standards converter modulators. |
|
27th Sep 2017, 4:26 pm | #9 |
Dekatron
Join Date: Jun 2005
Location: Warnham, West Sussex. 10 miles south of DORKING.
Posts: 9,147
|
Re: The pre-war picture on the screen
The 5" picture was based on the photographic 'N' print that was very popular during the 30s,/40s and 50s. It related almost exactly with the size of the picture on the 904.
We appear to have gone full circle with guys watching films and sport events on smart phones etc often with a screen size no bigger than the 904! I have no doubt that the large ion burn was due to the very low EHT voltage. Deflection of the beam would have been much softer allowing a larger area of coverage across the screen. The 904 is a heavy bit of kit! John. |
27th Sep 2017, 4:48 pm | #10 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
You can say that again!
There are three major hazards as I see it, working on this set. 1) Electrocution from the mains EHT - the bleeder resistor network is in plain view when you're working below decks. I try to keep it shrouded, but... 2) Inadvertently damaging or stressing the rare and fragile tube. This one is currently riding on a soft cushion covered in a Tesco bag! 3) Having your fingers' bones crushed beyond redemption, probably when your grip slips whilst heaving the chassis into another awkward working position. Meanwhile, I'm preparing to sweep the vision IF with my H-P sweep generator and a scope, but am being continually delayed by the set "taking off" at first warm-up. You name it, I have tried it, and I now fear "it's something to do with the tube". The effect is a pulsating bright orb on the screen, and only taking down the brightness (not contrast) and up again seems to clear it. The top cramping is now fixed (thanks Argus25) but it needed both the peaking resistor and my cap across the scan coils. Steve |
27th Sep 2017, 5:37 pm | #11 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
The problem could be an intermittent inter-electrode leak of some sort on the tube at warm-up. Altering the brightness and contrast now has no effect. I'm going to look for any visible effects on the scope at grid and cathode and try to deduce what's going on. Then I will repeat without the tube connected.
|
27th Sep 2017, 8:19 pm | #12 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
Yes, we have here a short-term grid-cathode leak on warm-up. The video on the scope at the cathode got flattened, but not with the tube disconnected. I confirmed this by putting an Avo grid-to-cathode on high R range and powering the tube heater from a bench supply. After the heater had finished warming up, the Avo moved from infinity over to the right.
Either it will get better, or it will get worse, or it can be lived with, or I may be looking for a 5FP4. What are my chances? Steve |
27th Sep 2017, 8:33 pm | #13 | |
Pentode
Join Date: Mar 2016
Location: Champaign, Illinois, USA.
Posts: 227
|
Re: The pre-war picture on the screen
Quote:
Until my red gun died I really enjoyed watching football (US) on my 1954 CT-100. I find B&W pictures more or less unwatchable except on old B&W movies. The ETF has a 60-line colour spiral-screw mechanical TV and it is very watchable! Of course, its a very high quality basic scanning system too. |
|
27th Sep 2017, 8:48 pm | #14 | |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
Quote:
Steve |
|
27th Sep 2017, 9:15 pm | #15 | |
Dekatron
Join Date: May 2010
Location: Greater Manchester, UK.
Posts: 18,713
|
Re: The pre-war picture on the screen
Quote:
Graham. G3ZVT |
|
28th Sep 2017, 12:22 pm | #16 |
Dekatron
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,535
|
Re: The pre-war picture on the screen
Yes, they were.
En(larged)prints. More expensive than contact prints of course...
__________________
....__________ ....|____||__|__\_____ .=.| _---\__|__|_---_|. .........O..Chris....O |
28th Sep 2017, 8:21 pm | #17 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
A quick update.
Using an arrangement that is in effect a wobbulator, I've now broadened the response curve whilst trying to keep it as smooth as possible. I've been rewarded with another set of gratings on the test card. But the focus has also changed. So I'm going to have a critical 'audit' of the focus chain resistors tomorrow with an eye to reducing drift. I like to imagine the pulsing effect of the wobbulator may have 'shocked' the tube out of its grid-cathode short. Anyway, it wasn't doing it this evening. Wishful thinking I expect... I'll know when it's cold tomorrow. Steve |
29th Sep 2017, 8:17 am | #18 |
Dekatron
Join Date: Jun 2005
Location: Warnham, West Sussex. 10 miles south of DORKING.
Posts: 9,147
|
Re: The pre-war picture on the screen
Just a point worth checking. Recently when working on a 905 the raster vanished leaving just a well focused uncontrolable dim trace. A very gentle tap on the 3/2 tube neck regained the picture. This is typical of an O/C cathode connection within the tube.
The actual emission was 100% but was degraded somewhat by a massive ion burn. The side wire connection socket was removed and the slide wires cleaned together with the tube socket connections. No improvement. A very careful examination of the lead out wires from the pinch proved they were all intact but a meter test between the lead out wires and the actual wire connections on the base proved the cathode wire to be O/C at the point where they were joined. The joint looked solid enough but very black. A quick clean and tin with the soldering iron completely cured the problem. Of course this fault could occur with and of the five connections. The late Gerry Wells did mention to me that these tubes did suffer from inter electrode shorts. John. |
29th Sep 2017, 9:26 am | #19 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
Thanks for the suggestion John.
Funny you mention this. In fact the cathode connection of this tube had earlier been found to be adrift and was carefully scraped and re-soldered, using snipe pliers as a heat shunt 'just in case'. The other connections have also been checked. The difference here is that this is definitely a leak or short occurring at warm-up, not an open connection. This has been confirmed (twice) with an ohm meter. I have also tried tapping the neck and rotating the whole tube. This didn't make any difference. |
10th Oct 2017, 10:14 am | #20 |
Nonode
Join Date: Feb 2005
Location: Worcestershire, UK.
Posts: 2,533
|
Re: The pre-war picture on the screen
Following help and advice from Argus25, a 5FP4 tube has now arrived from the Early Television Foundation in the States. All seems well, but it will be a short while before I can get started on the minor adaptations needed and try this.
As as also been pointed out, another thing I could try is to 'blow' the grid-cathode leak on the original 3/1 tube using a high voltage discharge from a capacitor. However, I understand this is not risk free. I shall try the 'new' tube first. Steve |