Featured
Latest
Share:
Notifications
Clear all

Forum 141

Stella ST8617U

87 Posts
11 Users
90 Likes
4,453 Views
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

Regarding V7, a PCF80 that I thought was meant to be an ECL80 according to the data, it is obvious that a PCF80 must be fitted as I have just spotted a receiver valve location chart inside the cabinet. ? 

It is bottom left, clearly marked....

IMG 5560 50

Interestingly, the EY86 inside the LOPT screening can is not indicated on the chart.

 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 15/03/2020 1:47 pm
sideband
(@sideband)
Posts: 4216
Famed Member Moderator
 

PCF80 was normally used in tuner units as mixer/oscillator but also found use as IF amplifier, video amp and (in some portables) as sound output. The ECL80 was normally used as sound output but could be found as sync separator, line osc and very oddly in at least one set a mixer oscillator in the tuner.

PCF80, ECL80, EF80 in particular were standard TV 'workhorse' valves. Nearly every TV of 50's vintage would have had at least one of these and in some sets more than one of each. In particular, the EF80 also found use in early FM radio's. There must have been millions of them made and there are still plenty of new ones about.

 
Posted : 15/03/2020 3:12 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
 

A derivative of the Stella ST8617U and the Philips 1768U is the model 1796U. This set employs the pentode section of a PCF80 as the line oscillator and the triode as the reactance control valve. Where Philips get clever is in the sync separator stage. An ECH83 triode-heptode is used instead of the ECL80. The triode functions as part of the frame oscillator, nothing too clever there, it's the heptode section where things really happen.

Normal grid current biasing is used, the positive going sync pulses are clamped to ground. The frame sync pulses are taken from the screen grids and the anode is supplied reference pulses from the grid of the line output valve.  The circuit functions as comparator, the coincidence between the sync pulses and reference pulses from the timebase results in a control voltage for the reactance valve. The sine wave oscillator is of the Hartley type. Later on the valve for this type of circuit was the PCF802.

Till Eulenspiegel.

 

 

 

 
Posted : 15/03/2020 4:39 pm
PYE625, PYE625 and PYE625 reacted
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

My later Stella ST2017U uses that circuit you have described Till.

https://www.radios-tv.co.uk/community/black-white-tvs/stella-st2017u/

 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 15/03/2020 6:12 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

Some of the new capacitors that will be making their way into the set arrived today. I obtained them from ebay and they are very economically priced, especially compared to NOS Mullard mustard types that when added up, would cost more than what I paid for the set !

I strongly suspect that many of the capacitor values are not too critical in the circuit of the set, but I want to replace the old capacitors with like for like.

IMG 5562 50

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 17/03/2020 5:40 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

I have replaced the bulk of the more critical capacitors and after cleaning the CRT this afternoon, I re-fitted it. There is still a fair bit of work to do, but the chassis is essentially working. The usual adjustments such as tuning, ion trap magnet position and the rear preset controls were required of course, but I am very pleased that there are no major issues. Sound is ok too. The tuner will need the contacts cleaned as it is rather intermittent and noisy.

IMG 5563 50
IMG 5566 50

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 20/03/2020 4:14 pm
Lloyd
(@lloyd)
Posts: 1897
Prominent Member Registered
 

That looks good so far! Is the CRT as good in real life as it looks in the picture?

Regards,

Lloyd

 
Posted : 20/03/2020 7:37 pm
Nuvistor
(@nuvistor)
Posts: 4594
Famed Member Registered
 

That looks great. ? 

Frank

 
Posted : 20/03/2020 8:21 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

Cheers Frank ? 

Hi Lloyd, the CRT is excellent. It has improved considerably over the space of an hour or so with good focus and a bright picture. I will later test the emission again on the B&K to see how much it has come up since I first tested it.

I have this evening cleaned the contacts in the tuner along with the biscuit terminals and this has vastly improved the tuner operation. I will re-assemble it to the control panel. I dismantled it so I could clean all the dust and grime away from the panel itself and all the fittings.

 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 20/03/2020 8:25 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

Here are a couple of pic's of the re-assembled tuner fitted to the side control panel. The CRT emission has risen a bit from when I first checked it. I need to check resistor values in the line hold circuit as the control is fully anti-clockwise to obtain lock. It drifts slightly and lock is right on the edge when the set is warm.

IMG 5567 50
IMG 5568 50
IMG 5575 50
IMG 5570 50

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 20/03/2020 10:17 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

I had trouble with getting the ion trap magnet to grip the neck of the CRT. Looking at the below first picture will reveal why. I sanded down the inner edge of the metal bracket and applied araldite to the broken portion of the flexible grip and inside the protecive sleeve. Hopefully, it will hold ok.

IMG 5576 50
IMG 5577 50

 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 21/03/2020 9:26 am
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

On a more serious note, there may be problems with the LOPT. Whilst I have had the overwind with 25vdc flowing through it over the past 6 days, it has dried quite nicely. In use in the set, it remains cool and EHT is approximately normal all the time. However, the width begins to creep in after 10 mins or so until about half an inch gap is at both sides of the picture. Now, the set has had a couple of hours use, but intermittantly. Never been on for longer than a few mins each time.

Upon examination of the LOPT after about 15 mins use, the main winding is very hot. I suspect the insulation to the earthed core is not good, so I am passing a voltage through the entire winding (about 38 ohms in total) in the hope that any moisture here may be driven out, if indeed this is the problem. I intend to leave the current flowing for a few days.

As can be seen from the below pictures, the overwind is quite seperate from the main windings so these have always been cool during the time I was drying out the overwind.

IMG 5578 50
IMG 5579 50
IMG 5580 50

 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 21/03/2020 4:17 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

I have used the Taylor 130 to test the insulation and can detect no leakage at 450v. This of course does not mean that there is no leakage at much higher voltages. I will give the process of warming the windings a chance and see what happens. If it makes no difference, then I will need to begin checking to see if any other conditions are changing, such as the line drive waveform for instance. I must admit to jumping to the conclusion that it is leakage in the LOPT causing the problem, but I could well be wrong.

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 21/03/2020 9:14 pm
Till Eulenspiegel
(@till)
Posts: 4947
Famed Member Registered
 

It may not be insulation failure that is causing your transformer characteristics to change over a given time.  Consider the possibility that the permeability of the ferrite core might be degraded with temperature rise. The condition is well known.

https://www.mag-inc.com/Design/Design-Guides/Designing-with-Magnetic-Cores-at-High-Temperatures

Till Eulenspiegel.

 
Posted : 22/03/2020 9:56 am
PYE625, PYE625 and PYE625 reacted
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

There are minor circuit changes (with reference to the Trader circuit in the library) as I mentioned earlier in the use of a PCF80 instead of an ECL80 for V7.

Aside from the base connection changes, R54 becomes 150k and is connected to the screen feed of the PL36, R59 is 180k and connected direct to HT instead of boost HT, and the grid resistor R60 of the PL36 is 2m2.

In this case, that 2m2 has risen to 3.5 megs. The PL36 data specifies 2m2 as the maximum limit for a grid resistor for line output applications only. I think it will be wise to replace this resistor as it is now rather high in value.

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 22/03/2020 11:44 am
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

In addition to R60 being high, R54 and 59 were also quite a bit higher than they should be. The effect of replacing these three resistors has brought the line hold lock quite a bit away from the control end stop. (The line hold control and R57 checked out ok).

When switched on from cold, the boost HT is near normal at approx. 565v. As the set warms, the boost will begin to slowly fall along with the width. Interestingly, it is now taking about 45 mins whereas it was around 10-15 mins before. Furthermore, the width only comes in about a quarter of an inch either side now. I didn't check the boost voltage before, so cannot say what it was when the width was less. Main HT is pretty much constant.

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 22/03/2020 8:39 pm
PYE625
(@pye625)
Posts: 5121
Famed Member Registered
Topic starter
 

Bit by bit I am able to leave the set on for longer before the boost HT drops. The width is coming in less and takes longer to do so too.

I really don't quite understand the mechanism behind this, but at a guess I would say either the LOPT or the scan coils are slowly giving up moisture or somehow leakage is becoming less. Any thoughts on the matter anyone? ? 

To understand the black art of electronics is to understand witchcraft. Andrew.

 
Posted : 23/03/2020 9:49 pm
Nuvistor
(@nuvistor)
Posts: 4594
Famed Member Registered
 

I think your assumption is more than likely the correct one. David often suggests running a low current through the LOPTX windings for a few days to dry it out.

 

 

Frank

 
Posted : 25/03/2020 9:44 am
Cathovisor
(@cathovisor)
Posts: 6391
Famed Member Registered
 

@nuvistor

Andrew's been doing that, Frank. With pictures to prove the point. 

 

 
Posted : 25/03/2020 9:48 am
Nuvistor
(@nuvistor)
Posts: 4594
Famed Member Registered
 

@cathovisor

oh dear, missed that. Some LOPTX’S were not that reliable 50-60 years ago, it’s a marvel that many still work after being in damp storage for many years.

 

 

Frank

 
Posted : 25/03/2020 10:04 am
Page 2 / 5
Share: