It is currently Sun Nov 19, 2017 9:48 am

Recreating the GEC 8161 Mirror Lid TV

Pre War 240/405, Post War 405 Line, B&W Dual Standard, Colour Dual Standard, B&W S/S 625, Colour S/S 625 line, Hybrids. Standards converters & modulators, video recorders.
 
Posts: 217
Joined: Mon Jun 04, 2012 11:40 am
Location: Newcastle on Tyne

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Ed Dinning » Tue Jul 07, 2015 7:14 pm

Hi Brian, just back from hols, let me know if you still need switch wafers, but PMor mail me with some details of types and shapes as there are many different types in my spares box.

Ed

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Tue Jul 07, 2015 7:54 pm

Thanks Ed. I'll take some pictures and send them to you. As an aside, what is the cardboard-like stuff used to wrap around transformers and scan coils called? It often carries tags or even voltage selection "buttons"? I think I am going to need some for the scan coils from my Marconi 707 TV. In this case, the stuff is wrapped round the yoke and carries the mounting brackets and connection tags. It is held tight around the yoke by corset-like lacing!

 
Posts: 217
Joined: Mon Jun 04, 2012 11:40 am
Location: Newcastle on Tyne

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Ed Dinning » Wed Jul 08, 2015 8:23 pm

Hi Brian, Presspahn I think; I also have some pretty thin and flexible SRBP here that might be useful.

Ed

 
Posts: 3781
Joined: Sun Feb 12, 2012 7:43 pm
Location: North Hykeham, Lincolnshire and Ilford, Essex (but not for much longer ...!)

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Terrykc » Thu Jul 09, 2015 11:37 am

peter scott wrote:... I'm guessing that the Aurora RF is double sideband ... If so then you really want a nice flat response from 3MHz up to 5.5MHz. If the Aurora had vestigial filtering then you might want a reduced response by 6db at 3MHz then flat from 3.75MHz to 5.5MHz to compensate for the doubling of power in the +/-0.75MHz region ...

Sorry, Peter, but double or vestigial sideband makes no difference! The receiver response must be the same for both - unless, of course, you have a dsb IF strip, which is clearly not the case here! In a practical design, the IF response will not stop dead at the vision IF frequency, so the IF strip will pass a vestige of the other sideband irrespective of the transmission characteristic

In the drawing below I have assumed the normal LO high which reverses the sound and vision carriers at IF whereas Brian's receiver has the LO below the signal frequency, so there is no inversion. However, this makes no difference to the explanation.

I have shown both DSB and VSB transmission with two IF responses. The first is the correct sloped VSB response whilst the second is an asymmetrical flat topped response. Note that the video output is the same for both, irrespective of the transmission type.

In each case, the vestige of the lower sideband is added to the main or only sideband, the output being the sum of both. If, as you were alluding to, both are received at 100% (as transmitted) the output below 750kHz will be double that of the higher frequencies.

DSB-VSB.png

 
Posts: 692
Joined: Sun Aug 07, 2011 8:48 am
Location: Edinburgh

Re: Recreating the GEC 8161 Mirror Lid TV

Post by peter scott » Thu Jul 09, 2015 12:35 pm

Hi Terry,

I do agree with you that where the IF response doesn't encompass the full lower sideband then the result would be the same regardless of DSB or VSB input signals. There will be higher signal power closer to the IF carrier and I was really just trying to interpret what Hawker says about shaping the IF response to equalise this.

I guess my first sentence should have read something like 0.5 to 5.5MHz rather than 3MHz to 5.5MHz.

Peter

hawker (Small).jpg

Hawker 2.jpg

 
Posts: 692
Joined: Sun Aug 07, 2011 8:48 am
Location: Edinburgh

Re: Recreating the GEC 8161 Mirror Lid TV

Post by peter scott » Thu Jul 09, 2015 1:31 pm

Image

Hi Terry,

Had Brian's set been receiving VSB he would actually be seeing the truncated sideband and not as in your drawing which is the reversed picture that results with an LO higher than the carrier.

Peter

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Thu Jul 09, 2015 4:22 pm

Terrykc wrote:If, as you were alluding to, both are received at 100% (as transmitted) the output below 750kHz will be double that of the higher frequencies.
As the Aurora outputs a double sideband signal, this must apply to all TVs manufactured after the introduction of VSB transmission!

The discussion has been enlightening and I think that I will go for a 5.5MHz centre frequency with a 3MHz bandwidth. The lower end (4MHz) should be as sharp a cutoff as possible but it will be a case of beggars can't be choosey!

 
Posts: 2843
Joined: Tue Jul 26, 2011 5:30 pm
Location: Gateshead

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Till Eulenspiegel » Thu Jul 09, 2015 4:44 pm

At least in the BT8161 there is that gap between the sound and vision passbands. Not so easy in the BT9121 because there doesn't appear to be any sound traps unless the 2.5Mc/s sound takeoff coil performs that function.

Till Eulenspiegel.

 
Posts: 692
Joined: Sun Aug 07, 2011 8:48 am
Location: Edinburgh

Re: Recreating the GEC 8161 Mirror Lid TV

Post by peter scott » Thu Jul 09, 2015 5:12 pm

In some of my earlier posts I was assuming Hawker's lists of IF frequencies were correct in saying that the 8161/8121 used the same 3.05MHz and 0.45MHz IF frequencies as the 3701/3702 and an LO frequency of 41.95MHz. This would have caused the sound to fold into the lower vision sideband so I guess that forced the 3701/3702 to only use the USB for vision.

With the 8161 service info showing a 41MHz LO then this fold over problem doesn't occur but did GEC continue to use only the USB or did they use both sidebands? It would give them more signal power but a need for more selective filtering to avoid sound on vision.

Peter

 
Posts: 2843
Joined: Tue Jul 26, 2011 5:30 pm
Location: Gateshead

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Till Eulenspiegel » Fri Jul 10, 2015 11:22 am

Looking at the RF circuits of the BT9121 it can be seen that the 2.5Mc/s sound take-off is in the anode circuit of the X41 frequency changer.
The primary of the first sound IF transformer is in series with the first vision IF transformer and will serve as the sound take-off and also as the sound trap for the vision passband. The BT9121 operates the on the lower vision sideband and I think that it can be agreed that the the BT8161 operates on the upper sideband.

Till Eulenspiegel.

 
Posts: 2843
Joined: Tue Jul 26, 2011 5:30 pm
Location: Gateshead

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Till Eulenspiegel » Fri Jul 10, 2015 6:03 pm

BT9121 update. The receiver chassis is back on the bench. I've determined that the frequency coverage of the three stage IF amplifier is in fact sensibly flat between 4 to 8.5Mc/s. Roll off starts at these frequencies. RF signals were injected into the control grid of the X41 hexode section.
The frequency changer is preceded by a two stage RF amplifier, two high slope KTZ41 RF beam tetrodes.

Till Eulenspiegel.

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Sun Jul 12, 2015 1:11 pm

Following the relatively in-depth discussions about the IF frequency and sidebands, I decided to make a couple of coils based on a centre frequency of 5.5MHz and using the upper sidebands. It's very strange but after winding, I tested the coils for resonance and they were all resonating much too low - around 2MHz. Now I had done all the calculations using a website with calculators embedded in it. This one:
http://www.midnightscience.com/formulas ... ators.html
The inductance calculator seemed to work as the coils were in the 30uH range! Anyway, I decided to install them anyway and see what the result was a bit encouraging! I fed the output of my video sweep generator into the Aurora. The result was this:
Video-Sweep-new-IFT5-&-Det-driver.jpg
The markers are 1, 1.5, 2 & 3MHz
It looks as though I'm getting somewhere so I will carry on up the strip - the detector driver and IFT5 have received the treatment so far. It's still deaf but this is probably because of the wide spread of transformer characteristics (say!).

 
Posts: 692
Joined: Sun Aug 07, 2011 8:48 am
Location: Edinburgh

Re: Recreating the GEC 8161 Mirror Lid TV

Post by peter scott » Sun Jul 12, 2015 8:54 pm

Hi Brian,

Given that this is a lin,lin plot (and probably looks worse than a log,log plot) I wonder what flatness you'd get from commercial IF strips in general from the 405 line era?

Peter

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Sun Jul 12, 2015 10:01 pm

It looks like -12dB at 3Mhz with a slight peak of -6dB at 2MHz so not so bad, I suppose! It will be interesting to see how things change as I work backwards towards the frequency changer.

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Mon Jul 13, 2015 9:37 am

Spoil Sport! :ccb. Maybe I should use the 525 setting! :qq1

 
Posts: 2843
Joined: Tue Jul 26, 2011 5:30 pm
Location: Gateshead

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Till Eulenspiegel » Mon Jul 27, 2015 4:55 pm

Confirming the details of the final vision IF transformer in the BT9121. The two outer windings have 54 turns each and the middle winding has 25 turns. The middle winding is the anode coil of the KTZ41, damped by the 3.3Kohm resistor. The outer windings are spaced apart by 10.5mm. Former diameter is 29mm.

Till Eulenspiegel.
Attachments
BT9121_0216.jpg

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Mon Jul 27, 2015 5:28 pm

Thanks for that, David.

 
Posts: 2843
Joined: Tue Jul 26, 2011 5:30 pm
Location: Gateshead

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Till Eulenspiegel » Thu Aug 06, 2015 8:16 pm

During the clear out of the room at the room at the back of the shop I found a photo copy of the original Baird T5 circuit diagram.
The T5 like the GEC BT8121 is a superhet and is also likely to have a very low vision IF. A Mullard 2D4A double diode is employed as a full wave vision demodulator. The T5 does not have an RF amplifier stage. The dipole aerial connects via the tuning coils to the grid of the Mullard TSP4 frequency changer. A triode valve is employed as the local oscillator, the LO is injected in the cathode circuit of the mixer.
Four TSP4 pentodes are used in the vision IF amplifier.
The diagram is dated March 1936.

Till Eulenspiegel.

 
Posts: 692
Joined: Sun Aug 07, 2011 8:48 am
Location: Edinburgh

Re: Recreating the GEC 8161 Mirror Lid TV

Post by peter scott » Wed Aug 12, 2015 6:39 pm

peter scott wrote:Hi Brian,

Given that this is a lin,lin plot (and probably looks worse than a log,log plot) I wonder what flatness you'd get from commercial IF strips in general from the 405 line era?

Peter


Out of curiousity I connected the attenuated tracking generator output of my TF2370 to the aerial of my HMV901. I removed the D42 (V7) detector and connected the spectrum analyser via an old HP scope probe and a little high impedance buffer (based on the Marconi TK2374 zero loss probe) to the anode connection.

moz-screenshot-3.jpg


P1020239 (Large) (Medium).jpg


The result looks like 3dB flatness and bandwith of 2MHz.

(The line at 42.5MHz is just the sound L.O. somewhat off tuned.)

Peter

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Tue Aug 25, 2015 1:14 pm

Only 2MHz, Peter! Maybe I am aiming too high :) . Thanks for the experimental results.
I have managed to get the esteemed Mr. M. Cox of Cox Box fame (esteemed in the broadcast industry, anyway!) to give me some guidance on the IF transformer design, so hopefully there will be some concentrated development - not just one candle with two ends but with 4 or 5 ends as it is at the moment.
Meanwhile, I have got the line output transformers back from Mike Barker. He disembowelled the LOPT from Till's BT9121 in order to copy it and a right job it was, said Mike. It was potted in pitch (bitumen) and had metal brackets to hold the unit centrally in the can. Having dissected the pattern, it was rewound using modern materials and insulation so it should be better than new. As for my unit, I had made a can of the same size but it opened in a different manner (I had not seen the original when I made the can) which provided a problem or two. The metal brackets and Paxolin bobbin had to be made and lams procured. Of course, lams of the exact size were not available so every lam had to be cropped to the required size. Anyway, the transformer looks the part and I am in the process of painting both mine and Till's two (line and frame) and will probably be trying them both during the next week.

 
Posts: 57
Joined: Sun Mar 30, 2014 6:23 pm
Location: Eindhoven, NL

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Jac Janssen » Tue Aug 25, 2015 7:38 pm

Brian,

Some pictures of the line output transformer in various stages of reproduction would be appreciated!

Jac

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Tue Aug 25, 2015 10:43 pm

If Mike (Murphymad) reads this, perhaps he could oblige with pictures. I would also appreciate a few for my project picture archive!

 
Posts: 1931
Joined: Mon Jul 25, 2011 12:12 pm
Location: Burghfield, Reading, UK

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Brianc » Mon Sep 21, 2015 12:19 pm

Mike has very kindly sent me some pictures of the original transformer during dissection. Here are some of them:
HPIM2527 (Medium).jpg
HPIM2532 (Medium).jpg
HPIM2528 (Medium).jpg
It doesn't look too special, in fact Mike said that it looks very like a high quality audio Xfmr of the period. However, the thick Paxolin bobbin and the spacing between the windings and the core suggest high voltages in use. There are also metal brackets to ensure that the assembly stays centrally in its metal can.

On another subject, it's now nearly three years since I started this project and its thread so I need to get my skates on - I don't even have the excuse of the LOPT problems!

 
Posts: 57
Joined: Sun Mar 30, 2014 6:23 pm
Location: Eindhoven, NL

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Jac Janssen » Mon Sep 21, 2015 2:20 pm

Thanks for the pictures Brian (and Mike).

All very interesting.

Jac

 
Posts: 2843
Joined: Tue Jul 26, 2011 5:30 pm
Location: Gateshead

Re: Recreating the GEC 8161 Mirror Lid TV

Post by Till Eulenspiegel » Mon Jan 18, 2016 11:01 pm

Hi Brian,
With regard to the low vision IF frequency of 4Mhz employed in the BT8161. We spoke about the Rediffusion TDUK2 cable TV system. In the 405 days the BBC was on a 4.95Mhz vision carrier and ITV was 8.95Mhz carrier. The Tete Beche system in which the BBC was on the upper sideband and ITV on the lower sideband. With such a low carrier frequency a full wave vision demodulator was desirable.

Till Eulenspiegel.

PreviousNext

Return to Television



Who is online

Users browsing this forum: No registered users and 2 guests