Alba T656 - UK Vintage Radio Repair and Restoration Discussion Forum

Colourstar · 10th Dec 2017, 2:13 pm

At least I think that's the correct model! There was no back cover when I acquired this set from the local outdoor junk market a few months ago. I used Jon's ever-reliable Valvepage to identify it.

This chassis uses two plug-in pcbs (timebase and IF) to allow for speedy servicing - or panel swapping- in the field. The sets were sufficiently popular for Practical Television to devote a service article to them across the July and July 1963 issues. A previous repairer obviously only had very large resistors to hand and several of these populate the print side of the pcbs.

Initially I could only get a very faint line whistle, then several power-ups later and seemingly of it's own accord the EY86 lit and a nice bright raster appeared. This remained consistent for sufficient time for me to carry out the usual component replacements to make the frame linear and obtain a very good picture on the Mullard AW43-80.

Now I've lost EHT again and there is only a very slight line whistle, detectable when turning the horizontal hold control. There is the correct -30v line drive on pin2 of the PL81, whereas the screen is about 50v higher than the 140v specified, running at around 195v. I'd be really grateful for any suggestions as to what else I should be checking. I have tried another EY86 to no avail.

Steve

1100 man · 10th Dec 2017, 3:13 pm

Hi Steve,
That's a very random chassis layout with the pcb stuck out the top! I assume it dates from 57/58 era when manufacturers were dipping their toes into using pcb's. Nice edge connector just to make it super unreliable! I bet print burn up's were common once the board got a good coating of nicotine

First suspect would be the boost cap and scan coupling cap if you haven't already changed them. Also there is probably a tuning cap across the windings somewhere in the few hundred PF region.
Do you have the circuit? If so, it might be worth posting the line timebase section so we can have a look.
All the best
Nick

Heatercathodeshort · 10th Dec 2017, 5:48 pm

Just a memory from the past. the EHT lead used to get trapped around the heater thermistor. It melted through and shorted out the EHT. Looks like you have a 655. The 656 was a 'transportable'. Same chassis.

You will have to check the voltages around the PL81 when the fault occurs. Either the drive is missing or maybe the screen volts. Dry joints and intermittent connectors were common. J.

Hunts smoothing bomb · 10th Dec 2017, 7:01 pm

I think the quoted voltages are during the fault condition John, have you tried changing the PL81 Steve?
Also are their any components in the cathode of the PL81?

Cheers

Colourstar · 10th Dec 2017, 7:03 pm

Hi Nick and John

The voltages mentioned in the first post are taken with the fault condition, so there is line drive (correct) and screen voltage (too high) present.

I can certainly see how the EHT lead could get pushed up against the heater thermistor, but it's OK in this case. Despite the plug-in boards, some aspects of this chassis are a pain to get at; notably the components on the bases of the LOPT valves which are hidden right behind the bracket holding the speaker. The bracket can be undone and slid up or down a bit, but can't be removed without considerable chassis dismantling. Not ideal.

Dodgy edge connectors and lifting print aside, the service article by LLJ does mention the tuning cap as a possible culprit for this problem. The boost cap has been changed.

Circuit attached although I'm not sure how readable it will be...

EDIT: Post crossed with Lee. Hi Lee! Yes, the PL81 has been changed to no avail.

Steve

Focus Diode · 10th Dec 2017, 8:34 pm

Seeing it was possible to get the set working following a few switch on's seems to suggest a dodgy contact somewhere. If the PL81 is cold in the fault condition first suspect would be the screen grid feed resistor.

From the burn up on the printed panel I wouldn't be surprised if there's a hairline fracture somewhere.

Cheers
Brian

Nuvistor · 10th Dec 2017, 9:20 pm

Same type of thing as a Brian suggests, where are you measuring the screen volts? If on the resistor, then a break in the circuit from resistor to valve base, if on the valve base, a broken connector on the valve base, especially relevant if it’s a McMurdo valve base.

Edit. A break in the anode circuit of the PL81 would cause the screen volts to drop. A break in the cathode circuit would remove the negative 30 volts on the grid.

FERNSEH · 10th Dec 2017, 11:20 pm

Alba models T655 and T656 employed two printed circuit boards which the dealer could repair or send back to the manufactures service department for repair or replacement. Was known as Alba's "packaged service".
Remove the anode connector of the EY86 and see if you can draw a flame like spark. If no or weak spark then remove the top cap connector of the PY81, if a healthy spark appears at the EY86 anode connector, then that will prove the boost capacitor has gone s/c.

DFWB.

FERNSEH · 10th Dec 2017, 11:55 pm

Circuit diagram of the line timebase. Very economical circuit design. Check that the anode load resistors R60 and R60 of the line oscillator are within tolerance.

DFWB.

Colourstar · 12th Dec 2017, 10:22 pm

Thanks for all the good suggestions. I'm beginning to suspect that the current fault is self-inflicted. I reckon I made a boob when replacing a couple of caps, c50 and C51 around the PL81 and PY81 as despite it's earlier reluctance, the set did seem to be working reliably before that. There are a couple of tag strips with quite a few components strung between them and I reckon I've connected something incorrectly. It's a congested area with limited access (at least that's my excuse).

I think the best thing to do is to make a sketch of the component layout on the set, mark the various values on it and check it against the circuit diagram. Hopefully then the error will become apparent.

mark pirate · 13th Dec 2017, 9:54 am

Quote:

I reckon I've connected something incorrectly. It's a congested area with limited access (at least that's my excuse).

I am afraid we all do this at some point, no matter how careful we are.
I have got into the habit of taking loads of close up pictures before I start restoring a set.

This has turned out to be a great idea, twice I have managed to spot my mistakes of soldering a component lead to the wrong tag

This usually happens due to an interruption such as a phone call, or just plain tiredness.

Mark

10th Dec 2017, 2:13 pm	#1
Colourstar Octode Join Date: Feb 2006 Location: Ilkeston, Derbyshire, UK. Posts: 1,398	Alba T656 At least I think that's the correct model! There was no back cover when I acquired this set from the local outdoor junk market a few months ago. I used Jon's ever-reliable Valvepage to identify it. This chassis uses two plug-in pcbs (timebase and IF) to allow for speedy servicing - or panel swapping- in the field. The sets were sufficiently popular for Practical Television to devote a service article to them across the July and July 1963 issues. A previous repairer obviously only had very large resistors to hand and several of these populate the print side of the pcbs. Initially I could only get a very faint line whistle, then several power-ups later and seemingly of it's own accord the EY86 lit and a nice bright raster appeared. This remained consistent for sufficient time for me to carry out the usual component replacements to make the frame linear and obtain a very good picture on the Mullard AW43-80. Now I've lost EHT again and there is only a very slight line whistle, detectable when turning the horizontal hold control. There is the correct -30v line drive on pin2 of the PL81, whereas the screen is about 50v higher than the 140v specified, running at around 195v. I'd be really grateful for any suggestions as to what else I should be checking. I have tried another EY86 to no avail. Steve Attached Thumbnails

10th Dec 2017, 3:13 pm	#2
1100 man Octode Join Date: Apr 2016 Location: Ventnor, Isle of Wight, & Great Dunmow, Essex, UK. Posts: 1,377	Re: Alba T656 Hi Steve, That's a very random chassis layout with the pcb stuck out the top! I assume it dates from 57/58 era when manufacturers were dipping their toes into using pcb's. Nice edge connector just to make it super unreliable! I bet print burn up's were common once the board got a good coating of nicotine First suspect would be the boost cap and scan coupling cap if you haven't already changed them. Also there is probably a tuning cap across the windings somewhere in the few hundred PF region. Do you have the circuit? If so, it might be worth posting the line timebase section so we can have a look. All the best Nick

10th Dec 2017, 5:48 pm	#3
Heatercathodeshort Dekatron Join Date: Jun 2005 Location: Warnham, West Sussex. 10 miles south of DORKING. Posts: 9,147	Re: Alba T656 Just a memory from the past. the EHT lead used to get trapped around the heater thermistor. It melted through and shorted out the EHT. Looks like you have a 655. The 656 was a 'transportable'. Same chassis. You will have to check the voltages around the PL81 when the fault occurs. Either the drive is missing or maybe the screen volts. Dry joints and intermittent connectors were common. J.

10th Dec 2017, 7:01 pm	#4
Hunts smoothing bomb Octode Join Date: Oct 2005 Location: Wimborne, Dorset, UK. Posts: 1,407	Re: Alba T656 I think the quoted voltages are during the fault condition John, have you tried changing the PL81 Steve? Also are their any components in the cathode of the PL81? Cheers __________________ Lee

10th Dec 2017, 7:03 pm	#5
Colourstar Octode Join Date: Feb 2006 Location: Ilkeston, Derbyshire, UK. Posts: 1,398	Re: Alba T656 Hi Nick and John The voltages mentioned in the first post are taken with the fault condition, so there is line drive (correct) and screen voltage (too high) present. I can certainly see how the EHT lead could get pushed up against the heater thermistor, but it's OK in this case. Despite the plug-in boards, some aspects of this chassis are a pain to get at; notably the components on the bases of the LOPT valves which are hidden right behind the bracket holding the speaker. The bracket can be undone and slid up or down a bit, but can't be removed without considerable chassis dismantling. Not ideal. Dodgy edge connectors and lifting print aside, the service article by LLJ does mention the tuning cap as a possible culprit for this problem. The boost cap has been changed. Circuit attached although I'm not sure how readable it will be... EDIT: Post crossed with Lee. Hi Lee! Yes, the PL81 has been changed to no avail. Steve Attached Thumbnails

10th Dec 2017, 8:34 pm	#6
Focus Diode Octode Join Date: Jan 2006 Location: Borough of Gateshead, UK. Posts: 1,420	Re: Alba T656 Seeing it was possible to get the set working following a few switch on's seems to suggest a dodgy contact somewhere. If the PL81 is cold in the fault condition first suspect would be the screen grid feed resistor. From the burn up on the printed panel I wouldn't be surprised if there's a hairline fracture somewhere. Cheers Brian Last edited by Focus Diode; 10th Dec 2017 at 8:41 pm.

10th Dec 2017, 9:20 pm	#7
Nuvistor Dekatron Join Date: Aug 2013 Location: Wigan, Greater Manchester, UK. Posts: 9,433	Re: Alba T656 Same type of thing as a Brian suggests, where are you measuring the screen volts? If on the resistor, then a break in the circuit from resistor to valve base, if on the valve base, a broken connector on the valve base, especially relevant if it’s a McMurdo valve base. Edit. A break in the anode circuit of the PL81 would cause the screen volts to drop. A break in the cathode circuit would remove the negative 30 volts on the grid. __________________ Frank Last edited by Nuvistor; 10th Dec 2017 at 9:24 pm. Reason: More info

10th Dec 2017, 11:20 pm	#8
FERNSEH Dekatron Join Date: Mar 2005 Location: Gateshead, Tyne and Wear, UK. Posts: 7,444	Re: Alba T656 Alba models T655 and T656 employed two printed circuit boards which the dealer could repair or send back to the manufactures service department for repair or replacement. Was known as Alba's "packaged service". Remove the anode connector of the EY86 and see if you can draw a flame like spark. If no or weak spark then remove the top cap connector of the PY81, if a healthy spark appears at the EY86 anode connector, then that will prove the boost capacitor has gone s/c. DFWB.

10th Dec 2017, 11:55 pm	#9
FERNSEH Dekatron Join Date: Mar 2005 Location: Gateshead, Tyne and Wear, UK. Posts: 7,444	Re: Alba T656 Circuit diagram of the line timebase. Very economical circuit design. Check that the anode load resistors R60 and R60 of the line oscillator are within tolerance. DFWB. Attached Thumbnails Last edited by FERNSEH; 11th Dec 2017 at 12:01 am.

12th Dec 2017, 10:22 pm	#10
Colourstar Octode Join Date: Feb 2006 Location: Ilkeston, Derbyshire, UK. Posts: 1,398	Re: Alba T656 Thanks for all the good suggestions. I'm beginning to suspect that the current fault is self-inflicted. I reckon I made a boob when replacing a couple of caps, c50 and C51 around the PL81 and PY81 as despite it's earlier reluctance, the set did seem to be working reliably before that. There are a couple of tag strips with quite a few components strung between them and I reckon I've connected something incorrectly. It's a congested area with limited access (at least that's my excuse). I think the best thing to do is to make a sketch of the component layout on the set, mark the various values on it and check it against the circuit diagram. Hopefully then the error will become apparent.