UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > Specific Vintage Equipment > Vintage Audio (record players, hi-fi etc)

Notices

Vintage Audio (record players, hi-fi etc) Amplifiers, speakers, gramophones and other audio equipment.

Reply
 
Thread Tools
Old 21st Feb 2017, 11:15 pm   #181
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Chris, they'll discharge regardless because the rest of the smoothers are effectively in parallel through the inductors - I've never heard of putting resistors in a power supply like this as the end user would not be expected to touch any of it. The valves will discharge the whole lot before they cool after switch off and would only not do so if there were a certain fault condition. From now on, every time someone posts with a vintage valve amplifier, we'd better tell them not to touch anything until they've fitted resistors across the smoothers in case of an unlikely fault.
Techman is offline   Reply With Quote
Old 21st Feb 2017, 11:21 pm   #182
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

I ought to add that you've probably noticed I keep telling the OP to make sure the capacitors are discharged before touching any of the connections. This is because of the very unlikely condition of a fault that could leave them charged, but more likely that if the unit is turned on and off quickly before the valves have fully warmed up, then this could leave the capacitors charged - but then this could happen with ANY amplifier.
Techman is offline   Reply With Quote
Old 21st Feb 2017, 11:50 pm   #183
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Just for reference, here's the circuit of the Dynatron version of the same amplifier from 1948, and as you'll see, no series capacitors and NO resistors across them.

However, one little 'mod' I would perhaps advise that you could do, and that's to include the 500mA fuse protecting the HT winding of the transformer. This was a modification by Dynatron on this later circuit, as the earlier one from 1945 does not include this fuse, so I guess they'd had the odd transformer disaster with the earlier amplifier design in the event of an HT short.
Attached Thumbnails
Click image for larger version

Name:	Diagram_5.jpg
Views:	52
Size:	91.7 KB
ID:	138016  
Techman is offline   Reply With Quote
Old 22nd Feb 2017, 12:19 am   #184
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 3,499
Default Re: Restore a Beau Decca

The Decola I have has series capacitors but no balance resistors and that has a much higher HT voltage! Safety was not high on the agenda in those days although I have seen resistors employed in later professional kit. The resistor was only a suggestion and would seem to be a good use of resistors that have already been purchased. Still it's your call, the only thing I would add is don't try discharging the capacitors using a short.

I am not convinced the fuse is intended to protect the transformer as it is unlikely it could deliver 500mA at 300V but it may help protect the user in the event the transformer insulation fails.
PJL is offline   Reply With Quote
Old 22nd Feb 2017, 12:33 am   #185
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

That's an interesting point about whether the fuse would blow in the event of a short, I personally think it would, although I could well be wrong. I've had a fault with that type of rectifier where the two anodes have shorted together and taken the transformer out. Unfortunately, no fuse of any rating in that particular position would have protected against that particular fault.

I've actually got the early version of that Dynatron amplifier and it hasn't got the fuse fitted. I keep meaning to fit the 500mA fuse modification, but never seem to get round to it.
Techman is offline   Reply With Quote
Old 22nd Feb 2017, 12:35 am   #186
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Quote:
Originally Posted by PJL View Post
The Decola I have has series capacitors but no balance resistors and that has a much higher HT voltage!
I'm very surprised about that. Are they electrolytic types?
Techman is offline   Reply With Quote
Old 22nd Feb 2017, 11:44 am   #187
thejazzageuk
Pentode
 
Join Date: Feb 2014
Location: Narberth, Wales, UK.
Posts: 171
Default Re: Restore a Beau Decca

Thanks Techman, PJL and Herald1360. Lots of good advice. In the end I installed the capacitors without the resistors or fuse, planning to add these later (if I get round to it).

So.. with the new caps in place and lamp limiter on, much better, still hum about one third level of music. Then I took out the lamp limiter, hum much lower frequency and much reduced volume, and reduces with volume increase. I then operated the middle Knob on the amp (radio/Gram/Brilliance), and hey presto- the hum has almost completely disappeared. Sounds fabulous! Any advice on how long I should let it run/ checks to do?

When the cleaning solution arrives, I will clean the valve contacts and socket contacts and perform the checks suggested by Techman in post No. 171.
Attached Thumbnails
Click image for larger version

Name:	newCapsInstalled.jpg
Views:	49
Size:	74.4 KB
ID:	138029  
thejazzageuk is offline   Reply With Quote
Old 22nd Feb 2017, 7:29 pm   #188
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 3,499
Default Re: Restore a Beau Decca

What we want to check is the output valves are reasonably well balanced. Measure the voltage across R47 and R48 again but be extra careful as the HT voltage will be much higher. Then measure the voltage from chassis to the anode of the output valves.

The voltage across R47/R48 should be about 2.4V. If either reads above 3V switch off.

Valve amplifiers fail because the output valve overheats and thermally runs away increasing the anode current until something gives way. Your amplifier uses automatic cathode bias where a resistor is placed in the cathode path (R44 and a bit from R45 and R46) and the grids are maintained near zero so that the valve finds a point of equilibrium.

This design shares the bias resistor between both output valves. R44 is chosen to set the equilibrium point correctly if both valves are passing the same current. If one valve is weak the shared bias arrangement results in an equilibrium point where the good valve is conducting far more current than designed for.

This is why it is important to check the anode currents. Being a bit paranoid, I am about to install some cheap Chinese 7 segment volt meters to monitor the cathode voltage on my Beam Echo kit. These amps do not have common cathode bias so the cathode voltage will tell me all I need to know.
PJL is offline   Reply With Quote
Old 23rd Feb 2017, 11:01 am   #189
thejazzageuk
Pentode
 
Join Date: Feb 2014
Location: Narberth, Wales, UK.
Posts: 171
Default Re: Restore a Beau Decca

Here are the results of the tests suggested by Techman:

What you should do is remove the valves, noting which is which and with your meter set on the lowest ohms range (200 ohms?) carefully and accurately measure the resistance of the heaters of each valve at the valve pins. Note and report these readings, also note the reading on your meter with the test leads shorted together and report this reading, as we’ll need to subtract this reading from the two readings of the heaters. At this sort of low resistance, the resistance of the meter and its leads etc. are important and depending on the meter could be around 0.3 to 0.4 ohms.


Resistance Valve A= 1.8 ohms
Resistance Valve B= 1.7 ohms
meter resistance = 0.8 ohms
So the dimmer valve does have a slightly lower resistance.

The table in the photo shows the results from the tests proposed by PJL.

I've also noticed that the cap can holding C47 and C48 is getting hot. Should I replace these straight away or do a few power cycles to see if it recovers?
Attached Thumbnails
Click image for larger version

Name:	table.jpg
Views:	22
Size:	24.3 KB
ID:	138082  
thejazzageuk is offline   Reply With Quote
Old 23rd Feb 2017, 4:50 pm   #190
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Quote:
Originally Posted by thejazzageuk View Post
So the dimmer valve does have a slightly lower resistance.
Are you sure about that? I would expect it to have a slightly higher resistance.

However, those resistance readings are not that far apart. One thing, though, those resistance readings are higher than they should be and I suspect the valves had just been run and were perhaps still warm when you took those readings. Ideally, the valves need to be stone cold when taking the readings as the resistance of the heaters increases considerably with temperature as they warm up. The other tests you've done do seem to now be showing that the fault is actually following the valve to the other position this time, so is a little more conclusive than it was before.

I saw your post earlier, so did some readings on a pair of valves myself, as I couldn't remember what they were from when I last did this test, pictures of these readings are below. The second valve in the picture is one that had a problem with a high resistance heater due to oxidised connections within the valve base pins, but has now been repaired. The third picture shows the meter with its test leads shorted together giving 0.3 ohms, a bit lower than yours. The forth picture shows the base of the previously faulty valve after I repaired it. I must stress the point that you have to make sure that the test leads make perfect connections to the valve base heater pins and this can only really be done with crock clips - you should invest in some good crock clip leads if you don't already have them. The three handed operation of trying to hold test probes onto valve pins is a disaster waiting to happen if the valve flips out of your hand and lands on the floor. Also, not being able to make a good contact, which doesn't matter with the 'high' resistance readings you've been making, but is very important with these 'low' resistance readings. To make a point, those 1.0 ohm readings both read 1.1 ohm until I 'jiggled' the crock clips to make a better contact - it really is that important and at these low resistances 0.1 of an ohm IS important. So with the meter resistance of 0.3 ohms taken off, the cold filaments of the PX4 valves in good order should be around 0.7 ohms. From the readings you've taken with your valves, the resistance of the filaments is coming out at about 1.0 ohm - a little high if they were stone cold with good contacts from the test leads.

Having said all of the above, I'm tending to now be leaning towards more of a difference in the actual normal emission of the valve due to the emissive coating of the filament rather than an actual heater connection problem, but I still think you need to clean up those valve pins, particularly the heater ones as they're taking an amp each, which is a high current for that type of connection. Also switch cleaner in the sockets of the valve bases on the amplifier chassis to properly rule out any bad connections there. The 'lower' valve could also be a little bit 'gassy'.

Rightly or wrongly, I'm personally not as concerned about this difference in the valves as PJL is, and I'm a bit wary of this getting 'over complicated', but we'll see what he thinks when he posts. It could be that the gram was operated for a long time with one of those coupling capacitors quite leaky that gave one of the valves a bit of stick, or it could even have been a bad heater connection in one of the valves which caused the other one to take more HT current. I think what's happened in the long distant past is that one of those balancing resistors failed open circuit. The difference in the, by then, slightly ageing electrolytics, caused one to get too much voltage across it, so it started to fail. This then caused more ripple further down the line which has given the double can electrolytic that's now getting hot as well, a hard time too. The previous owners probably carried on running it and put up with it 'humming' away until it finally got too much to hear the music over - just guessing, but I bet it was something like that!

I would guess that it's probably only one section of the 'double can' capacitor that's developed a bad leak and is getting hot, but I think it's doubtful that it will recover and will need disconnecting and two of your new 10uf replacements connecting in circuit in its place, just to be on the safe side.

Valve heater testing pictures below:-
Attached Thumbnails
Click image for larger version

Name:	P1100092.JPG
Views:	25
Size:	138.7 KB
ID:	138085   Click image for larger version

Name:	P1100093.JPG
Views:	26
Size:	150.1 KB
ID:	138086   Click image for larger version

Name:	P1100094.JPG
Views:	24
Size:	145.1 KB
ID:	138087   Click image for larger version

Name:	P1100095.JPG
Views:	22
Size:	159.7 KB
ID:	138088  

Last edited by Techman; 23rd Feb 2017 at 5:00 pm.
Techman is offline   Reply With Quote
Old 23rd Feb 2017, 5:36 pm   #191
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 3,499
Default Re: Restore a Beau Decca

This is without the lamp limiter? The voltage readings look better matched now. I expect you didn't leave them long enough to fully warm up as the sum of the currents is different depending on the socket they are placed in.

The caps need to be replaced as if they were going to recover they would have done that by now. The entire power for the output stage passes through these capacitors so again we need something with a ripple current rating of at least 300mA.

It looks like F&T have stopped their range of low value dual can high ripple current capacitors. These should be OK:
RS: 739-4582 Nichicon CA series UCA2W100MHD 10uF 450V 105degC

Decca must have been trying to save metal as cramming the electrolytics in such a small space under the chassis does not look too clever. The elevated temperatures would have reduced performance and shortened life. The one I have proposed is 105degC which will help.

Our posts crossed: I agree, it looks like the valves will be OK. Those caps need changing before it is safe to put it in the cabinet, then a further check as the HT will have changed, a soak for a few hours, final measurements and then it should be done. Any other ideas on a source of electrolytics? I would check and replace any out of tolerance resistors (well actually I would probably replace the lot rather than waste time checking them) - what do you think?

Last edited by PJL; 23rd Feb 2017 at 5:50 pm.
PJL is offline   Reply With Quote
Old 23rd Feb 2017, 10:26 pm   #192
thejazzageuk
Pentode
 
Join Date: Feb 2014
Location: Narberth, Wales, UK.
Posts: 171
Default Re: Restore a Beau Decca

Thanks Techman and PJL.

I will clean up the valve terminals and redo the readings.

For the capacitors- I think the can contains 1 16microF and 18microF. Are the ones I showed in post 143 suitable(2 in parallel for 16microF, and a single one for the 8)? The spec for these is here:
http://www.justradios.com/axialelectrolytics.html
thejazzageuk is offline   Reply With Quote
Old 24th Feb 2017, 12:26 am   #193
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Ah yes, I was forgetting that one of these was a 16uf rather than them both being 8uf. It would be better if the 16uf was replaced by a single capacitor, but there’s no real reason why you shouldn’t use two 8uf (or 10uf) capacitors in parallel seeing as you have them already – you can always change it later if you want. I see that company does the exact values the same as the originals. These that you have should be of very good quality due to where you’ve sourced them from and are 105c temperature rated, and although I take the point that PJL has made about ripple rating, I personally think that the brunt of the ripple has at this stage been absorbed by the paralleled 4 x 1uf reservoir, so the good quality electrolytic ones that you have should be fine – have a look at what’s written on the can of the original to see if a ‘ripple rating’ is noted on it as I’ve not normally come across high ripple capacitors being used in this position before. I see that the company you got your capacitors from do a 4uf 600 volt working electrolytic which I think you could probably have got away with using – well, I would have given it a try, but what you’ve got with the four that you’ve used we pretty much know will cover the ‘ripple’ aspect regardless, so stick with what you’ve got. Obviously, make sure you connect the electrolytic ones the correct way round.

As regards the resistors, a quick look at the circuit shows that just about all of them can be measured in circuit, so just run your meter probes across them all to check their value – power disconnected from the amplifier, of course. The only resistors that I’ve honestly ever had problems with are those anode load resistors R47 & R48 on your diagram. I’ve known them degrade with internal arcing and messing about causing continuous cracking in the speakers, obviously regardless of volume control setting. However, that doesn’t mean that yours are going to fail in this way and you’ve already confirmed that they’re of the correct value, so unless any of the other resistors read way off, then I’d leave the originals in place. If you get any way out of spec resistors, then report them for advice.

There are several other capacitors that we’ve talked about that just need to be thought about, but we’ll get to them when you’ve sorted that double can replacement – unless you’re running out of stock and intend putting an order in, then I would advise getting replacements for these others just in case, not forgetting that cathode de-coupler C36 that I think I mentioned as a possible a long time ago, but may actually not need replacing.

You’re getting on really well with this amplifier!

Last edited by Techman; 24th Feb 2017 at 12:31 am.
Techman is offline   Reply With Quote
Old 24th Feb 2017, 5:55 pm   #194
thejazzageuk
Pentode
 
Join Date: Feb 2014
Location: Narberth, Wales, UK.
Posts: 171
Default Re: Restore a Beau Decca

Thanks Techman. I've now replaced the Cap Can with just 2 capacitors- a 10 and a 22 microF, as I found I had the latter in the kit.

Still works, but I notice that there is some distortion with the music (only low level, but it wasn't there earlier- so am doing a bit of a substitution exercise to see if reverting back to the original caps loses this distortion.

I am using the 'brilliance' control in its full clockwise position, if I turn it to the centre 'gram' position there is still quite a bit of hum alongside the music. This may reduce as I start to use the amp but I will keep checking this as we go along.
I've attached a pic of the original cap (tried looking for a replacement as would have preferred to keep 'the look' but so far can not find one)- and the replacements in situ.
Attached Thumbnails
Click image for larger version

Name:	capOrig.JPG
Views:	36
Size:	31.9 KB
ID:	138157   Click image for larger version

Name:	new8 and 16.jpg
Views:	40
Size:	74.2 KB
ID:	138158  
thejazzageuk is offline   Reply With Quote
Old 24th Feb 2017, 6:29 pm   #195
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Looking at your picture it looks like the original capacitor sections are still electrically connected in circuit. I may be wrong about this but it just looks like they're connected. Make sure that NO electrical connection exists to the old 'double can' capacitor.

Just to add: It doesn't matter if the negative, can or chassis connection of the capacitor is still connected, it's just the two positive connections that MUST be completely isolated electrically.

Last edited by Techman; 24th Feb 2017 at 6:40 pm.
Techman is offline   Reply With Quote
Old 24th Feb 2017, 6:35 pm   #196
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Quote:
Originally Posted by thejazzageuk View Post
I am using the 'brilliance' control in its full clockwise position, if I turn it to the centre 'gram' position there is still quite a bit of hum alongside the music. This may reduce as I start to use the amp but I will keep checking this as we go along.
C38 and C43 are the other capacitors I mentioned yesterday that still may need looking at and could be responsible for hum. Also, a complete check of resistor values, as a badly out of value resistor may be responsible for distortion.
Techman is offline   Reply With Quote
Old 24th Feb 2017, 7:58 pm   #197
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Ok, this is important!

Once you've confirmed that there's no remaining electrical connections to the two positive terminals of that original double capacitor, I think it's time we had a list of HT voltage readings (full mains, no lamp limiter), but keep an eye on the 'can' temperature of those remaining electrolytic capacitors while you're making them.

So with your meter negative probe firmly connected to the chassis, and starting off with the meter set to its highest DC voltage range, take, note down and post the following readings:-

1) The hot end of the new reservoir set of four capacitors.
2) The positive end of C48.
3) The positive end of C47.
4) The positive end of C43.
5) The positive end of C38.
6) I would also like to see the anode voltages of the three preceding valves.
7) Also, the voltage on the screen grids of the two pentodes, ask if not sure.
Techman is offline   Reply With Quote
Old 24th Feb 2017, 8:43 pm   #198
PJL
Dekatron
 
Join Date: Sep 2005
Location: Seaford, East Sussex, UK.
Posts: 3,499
Default Re: Restore a Beau Decca

I would also recommend measuring the voltage across R47 and R48 again.

The larger can capacitors like this used in valve amplifier reservoir and smoothing positions typically have a ripple current rating of 300mA or more (pic is a Radford 5-20 amp). Lower ratings will mean higher ESR and greater dissipation in the capacitor. The rating of the new capacitors is an unknown but I would at least check they are not getting warm.
Attached Thumbnails
Click image for larger version

Name:	capacitor.jpg
Views:	32
Size:	22.1 KB
ID:	138166  
PJL is offline   Reply With Quote
Old 24th Feb 2017, 11:53 pm   #199
Techman
Octode
 
Techman's Avatar
 
Join Date: Nov 2009
Location: Lincolnshire, UK.
Posts: 1,424
Default Re: Restore a Beau Decca

Quote:
Originally Posted by PJL View Post
The larger can capacitors like this used in valve amplifier reservoir and smoothing positions typically have a ripple current rating of 300mA or more
Yes, but as you can see in your picture, this ONLY applies to the 'Red, Outer' reservoir part of the can. The 'Yellow, Inner' part is the smoother and has NO ripple rating after it. Ripple rating is not normally required once past the reservoir section of a power supply of this type.

I have a suspicion that those capacitors supplied by 'Just Radio' are actually good enough to handle ripple current. I suspect that it was just a problem in the past when 'some' capacitors had to be specially made to handle ripple current and were therefore always on the 'outside' of a 'multi-can' capacitor so as to aid cooling.

Other forum members 'in the know' may like to comment on these new, but high quality capacitors, as to whether this may be the case, although this is probably a topic for a separate discussion.
Techman is offline   Reply With Quote
Old 28th Feb 2017, 9:49 pm   #200
thejazzageuk
Pentode
 
Join Date: Feb 2014
Location: Narberth, Wales, UK.
Posts: 171
Default Re: Restore a Beau Decca

Here are the readings from the resistors that I could access relatively easily. I may have disturbed something though as now the hum is very loud- much louder than the distortion notices after replacing the cap can. I will keep looking out for a replacement 8 + 16 microF cap can to replace the existing one and keep the original look iff possible, and also will investigate and do some more substitution checks on the cap can too- to try to get back to the high audio quality experienced previously.
Attached Thumbnails
Click image for larger version

Name:	resistors.JPG
Views:	38
Size:	37.9 KB
ID:	138349  
thejazzageuk is offline   Reply With Quote
Reply

Thread Tools



All times are GMT. The time now is 3:27 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2017, vBulletin Solutions, Inc.
Copyright ©2002 - 2017, Paul Stenning.