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

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 11th Aug 2018, 2:03 pm   #21
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
Default Re: Valve construction.

Quote:
Originally Posted by kellys_eye View Post
Newbie question - does the size of the evacuated envelope have any bearing on the parameters of a valve?
G-J's covered it, but a larger envelope gives a larger reserve of vacuum to cope with outgassing. The envelope runs cooler so less risk of suck-ins and getter spray can be kept further from the mica plates etc.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 11th Aug 2018, 2:25 pm   #22
David Simpson
Nonode
 
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,839
Default Re: Valve construction.

GJ has raised a good point in mentioning clouds of electrons at the cathode, and a similar "cloud" at the anode. The cathode cloud's electrons soon mostly whizzes of to the anode when HT reaches full potential, but the anode cloud of electrons was seen as a problem as triodes, then SG(screen grid)valves, were developed in the 1920's. Hence a pentode's Grid3. G3, being connected to & at cathode(more or less -ve) potential, persuades the -ve electrons from being attracted to the screen's +ve potential, or in fact those random electrons encouraging grid current under certain biasing & signal conditions.
Back to mentioning the orientation of anodes & cathodes - have a look through The Valve Museum site. P1's, HL610's, LS2's & 3's :- very early valves, some with "Hat" type anodes, or horizontal anodes, not vertical tubular shapes. Researching those post WW1 & 1920's valves, is a good way of understanding the modern-ish valves we now take for granted in vintage pursuits.
By the way, those with AVO VCM's will often notice the slow dispersal of the cathode "cloud", after completion of testing. Reveals itself under C/H INS.

Regards, David
David Simpson is offline  
Old 11th Aug 2018, 9:01 pm   #23
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Valve construction.

Quote:
Originally Posted by G8HQP Dave View Post
You can probably get a very rough first approximation of triode mu by comparing g-k distance with a-k distance. Mu ~ Da-k/Dg-k. This is because what actually moves the electrons away from the cathode is the electric field they see there, and this is a linear combination of that created by the grid and that created by the anode
That's a really useful insight into the internals of a valve, Dave!

I guess the same formula could be used for a pentode's µg1-g2 (obviously with the g2-k distance substituted for your a-k)

Quote:
Originally Posted by David Simpson View Post
It confused me as a young teenager starting out in the RAF, then years later as an instructor was often asked by young trainee apprentices - why does the pictorial representation of a valve(and its current flow) appear to have vertically mounted electrodes when the beam of electrons within a valve actually(in most cases) flows horizontally?
Well, I have seen old circuit diagrams where the valve symbol appears with the cathode on the left, the grid as a broken vertical line, and the anode on the right! Which would hopefully please you. But I do prefer the cathode at the bottom, the grid in the middle, and the anode at the top - because it allows potentials generally to rise as you go up the diagram.

Quote:
Originally Posted by GrimJosef View Post
Actually the electrons at the cathode aren't (usually) nearly as 'warm' as they are when they arrive at the anode. Cathode materials are chosen for having a low work-function (a measure of how hard it is to pull electrons out of them). I seem to recall that the electrons in the space-charge cloud which blankets the cathode have a typical energy below 1eV or so. By the time they've been accelerated by the anode-cathode potential they will have picked up perhaps 300-400eV of energy which is then lost, as you say, by inelastic scattering at the metal anode, heating the anode up as a consequence.
That's as I understand it too - electrons liberated from the cathode have very little energy. If they had a lot, you could connect the anode of a diode to the cathode (externally) and observe a large current flow. But you get hardly anything. Or, you could connect a voltmeter externally and see a large negative voltage on the anode, as it 'caught' electrons whizzing away from the cathode, until it accumulated enough negative potential to just about repel any more. But you don't - you may see 1V at most, which is right in-line with G-J's assertion that the electrons have around 1eV of energy at most. (If the electrons had significant energy, you would also need extra heating power to compensate for the energy the electrons carry away as you draw more anode current).

It is the anode voltage accelerating the electrons which speeds them up - 250V on the anode means that each electron hitting it has 250eV of energy. And that's why the anode of the EL84 in your radio gets hotter than the anodes of the EZ80, even though the EZ80 carries more current than the EL84!
kalee20 is online now  
Old 11th Aug 2018, 10:33 pm   #24
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
Default Re: Valve construction.

You still have to pour a lot of heater watts into the EZ80 to be able to emit the required rate of electrons/second to make up the set's total HT current

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 11th Aug 2018, 10:43 pm   #25
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Valve construction.

You certainly do. Though, the EZ80's heater draws 600mA whereas the EL84 guzzles 760mA!

I'm thinking that the EZ80's anode (which does not get anything like so hot as the EL84's) may be bright internally, thus reflecting radiated heat back to the cathode, so it will need less power to keep it up to temperature. But the EL84 has an anode that dissipates 12W, so it has to be much bigger, plus it's farther away to make room for all those grids (which have to be kept coolish else they might start emitting themselves), so the EL84 cathode loses heat quite readily and thus needs more heater power.
kalee20 is online now  
Old 12th Aug 2018, 10:39 am   #26
David Simpson
Nonode
 
Join Date: Mar 2008
Location: Aberdeen, UK.
Posts: 2,839
Default Re: Valve construction.

Right enough, as kalee20 says in post 23, as well as the bog-standard orientation in most circuit diagrams, from the early days onwards there have been a number of different & sometimes weird representations of valves.
I take my hat off to Andy for spending time opening up valves & examining their electrodes. Something I've never done intentionally. Just when a valve has been accidentally damaged. Many valves which pass through the hands of vintage radio collectors are often well silvered & often grubby, so examination is limited.
I was lucky, even privileged, last year in being allowed to test & catalogue a BVWS chum's valuable collection of rare early valves, many NOS & still in their boxes. The glass envelope being clear & unblemished enabled me to see primitive electrode structures, which up till then I could only study on valve museum sites.
It just goes to show that, electronics wise, the learning curve still continues after 3 score years & ten. Even more wonder & enjoyment was experienced with these old valves when discovering inside some boxes - original graphs of expected Ia/Vg(gm) curves. Then - lo & behold, many of these valves were able to just about replicate their tabulated spec on a valve tester !

Regards, David
David Simpson is offline  
Old 12th Aug 2018, 11:14 am   #27
Herald1360
Dekatron
 
Herald1360's Avatar
 
Join Date: Feb 2007
Location: Leominster, Herefordshire, UK.
Posts: 16,528
Default Re: Valve construction.

Quote:
Originally Posted by kalee20 View Post
You certainly do. Though, the EZ80's heater draws 600mA whereas the EL84 guzzles 760mA!
Maybe the EZ80's anodes being closer to its cathode helps, too.
__________________
....__________
....|____||__|__\_____
.=.| _---\__|__|_---_|.
.........O..Chris....O
Herald1360 is offline  
Old 13th Aug 2018, 3:43 pm   #28
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve construction.

The EZ80 anode being closer to the cathode means that a given anode current flow requires a smaller anode voltage, so less voltage drop in the PSU and less heat at the anode.

It could be (I am not sure) that an EZ80 needs less heater power because it has a slightly smaller cathode surface area. Or it could be that there is less concern about maintaining a full healthy cathode space charge throughout the AC cycle. In an output valve a weak cathode space charge might mean some rounding of signal peaks; in a rectifier the same phenomenon could produce a useful reduction in peak current. Sadly, the people who might know have probably gone.
G8HQP Dave is offline  
Old 13th Aug 2018, 6:21 pm   #29
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,803
Default Re: Valve construction.

The little EZ80's cathode is having to emit every bit of current which passes through all the other valves in a set. And it runs on less than the sum of all their heater powers. And it's expected to run these currents with lower anode-cathode volts.

Oh, yeah, and it has to do it in narrower pulses than half a cycle, due to peak charging and the size of the reservoir.

Bit of an unsung hero.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is online now  
Old 15th Aug 2018, 12:42 pm   #30
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Valve construction.

Quote:
Originally Posted by G8HQP Dave View Post
The EZ80 anode being closer to the cathode means that a given anode current flow requires a smaller anode voltage, so less voltage drop in the PSU and less heat at the anode.
Yes - smaller Vak but smaller distance ak means same electric field strength, so same pulling power on space charge. Which translates to same Ia.

Quote:
Originally Posted by G8HQP Dave View Post
It could be (I am not sure) that an EZ80 needs less heater power because it has a slightly smaller cathode surface area. Or it could be that there is less concern about maintaining a full healthy cathode space charge throughout the AC cycle. In an output valve a weak cathode space charge might mean some rounding of signal peaks; in a rectifier the same phenomenon could produce a useful reduction in peak current. Sadly, the people who might know have probably gone.
I'd be a bit unsure of this too, though I'd suggest that no significant current limiting happens due to a weak space charge. Because if it did, Vak would rise, and heat generated at the anode would also increase. Also, the higher Vak would accelerate any positive gas ions towards to cathode, and with a depleted space charge there'd be little to neutralise them so they would continue to accelerate until they smashed into the cathode surface. And that's not good for the emissive layer
kalee20 is online now  
Old 15th Aug 2018, 2:27 pm   #31
G8HQP Dave
Rest in Peace
 
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
Default Re: Valve construction.

Yes, ion bombardment is not good.

The effect on anode heat might be less than expected, though. Basically you have a choice between a current peak (so boosting I) or a voltage peak (boosting V) - either will boost anode heat IV. I was speculating, so quite likely to be wrong!
G8HQP Dave is offline  
Old 22nd Aug 2018, 11:19 pm   #32
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Valve construction.

By coincidence I've had a surprising 'valve construction' experience today.

I have a preamplifier in for service which has been modified at least twice by other people in the past. I don't have a circuit for this model and even if I did I couldn't be sure about any subsequent mods. As part of the mystery there doesn't seem even to be a definitive published valve line-up.

All five of the valves were clearly B9A double triodes and having traced the circuit it was clear they all had the ECC81/2/3 pin connections. Since this was a stereo pre I could at least be certain that corresponding valves in opposite channels were the same. On this basis it seemed that in fact they were all ECC83s.

The first two pictures show the two valves from the phono stage, one from the left channel and one from the right. One is clearly marked ECC83 and although the other has had the end of its label wiped off it looks exactly the same as the first. The anodes are identical, as are the micas. All of the connections between the pins and the electrodes follow exactly the same paths. The getter rings match perfectly. Even the remaining printing on the second valve looks very much like the printing on the first.

Imagine my surprise then when I came to check the type codes. One of them is I6 which is perhaps the most common of the ECC83 codes. But as you can see from the third picture the other is Gf. So it's not an ECC83 at all. It's an ECC82. These valves are not similar. The '83 has a mu of 100, and with 250V a-k and just -2V on the grid the valve passes 1.2mA. The '82 has a mu of 17 and with the same Va-k the grid voltage can be taken all the way to -8.5V and the valve will still be passing 10.5mA.

I was so surprised I actually put the valves into my AVO CT-160. Sure enough the '83 tested just like an '83 and the '82 like an '82. To look at they are the absolute spitting image of one another. Peas in a pod. But they're very different valves.

The last picture (which is a blow-up of part of the second one) shows what seems to be the only visible difference. If you look into the anodes through the small circular holes about half way up the picture and close to its left and right edges, you can just make out the grid wire wrapped around the silver-coloured grid rod. In the valve on the left (the '82) the wire is thick and coarsely spaced. In the '83, on the right, it is thin and finely spaced. The grid-cathode gap might be different too, but it's essentially impossible to see the cathode so I can't tell.

This is how the valves were when the pre arrived here. The owner likes the sound. I wonder if it'll sound different to him when I've replaced the '82 in one channel with an '83 to match the other one ?

Cheers,

GJ
Attached Thumbnails
Click image for larger version

Name:	Concordant ECC83s 01 small.jpg
Views:	57
Size:	48.1 KB
ID:	167955   Click image for larger version

Name:	Concordant ECC83s 02 small.jpg
Views:	54
Size:	49.9 KB
ID:	167956   Click image for larger version

Name:	Concordant ECC83s 03 small.jpg
Views:	54
Size:	42.8 KB
ID:	167957   Click image for larger version

Name:	Concordant ECC83s 04.jpg
Views:	62
Size:	47.8 KB
ID:	167958  
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 23rd Aug 2018, 7:18 am   #33
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,639
Default Re: Valve construction.

Interesting GJ, this is why I started the thread. As you say you can clearly see the grid difference on your pic 3. I'd bet the cathodes are identical, at least from a first glance, but the ECC82's must be beefier, it has to "pass" more current, 10 times more.

I'm still ploughing through some the suggested reading so havn't commented for a while but am still reading the posts.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 23rd Aug 2018, 8:24 am   #34
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Valve construction.

I don't know what it is that limits the current in the ECC83 (the datasheet says it shouldn't be taken above 8mA, or the anode dissipation above 1W, whereas the '82 is rated for use up to 20mA and 2.75W). I might have guessed that in the '83 they were aiming to keep the grid temperature down, perhaps because the wire is so fine and so close to the cathode (if it is). But they haven't gone to the trouble of using copper grid rods for better conduction nor have they welded radiating fins onto the top of the grid rods, as you sometimes see in other valves.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 23rd Aug 2018, 8:51 am   #35
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Valve construction.

ECC82 and 83... yes I've been able to tell an unmarked one by the grid, but only if I have a pair of known ones to compare against (and that's ignoring the possibility the unmarked one is an '81).

It's a good point about anode dissipation, why can the '82 tolerate nearly 3 times as much heat at its anode when they have apparently the same size and shape? Can Synchrodyne help?
kalee20 is online now  
Old 23rd Aug 2018, 8:59 am   #36
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Valve construction.

I haven't thought about what really controls the grid temperature. There are both heat inputs (electron impact heating and radiation from the cathode and the anode) and heat outputs (conduction down the grid rods to the connectors and thence to the pins and radiation back to the other electrodes) to consider. Maybe the anode temperature has to be kept low not to protect the anode itself but to help manage the grid temperature ?

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 23rd Aug 2018, 9:27 am   #37
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Valve construction.

Keeping grid temperature down is important to minimise grid emission. If a valve is intended to be operable with super-high values of permitted Rgk, then it could be inferred that grid current has been thought about and carefully minimised. But Rgk is the same for both valves, indicating that grid currents are similar.

I'd expect most of the temperature rise of the grid to be due to the cathode anyway. In the high-mu ECC83 the grid is closer to the cathode and the anode is relatively distant (as per G8HQP Dave's post #10), on that basis you'd rather think the '83 could cope with MORE anode dissipation than the '82, not less!
kalee20 is online now  
Old 23rd Aug 2018, 9:41 am   #38
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Valve construction.

I wonder if the observed very fine grid wires associated with the ECC83 would be more prone to microphony when thermally expanded/changed in temper than the relatively thick ECC82 wires? Also perhaps given the application implication that the ECC83 will be used in circuits offering higher gain and with lower-level signals.
turretslug is offline  
Old 23rd Aug 2018, 10:28 am   #39
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: Valve construction.

Quote:
Originally Posted by kalee20 View Post
... I'd expect most of the temperature rise of the grid to be due to the cathode anyway. In the high-mu ECC83 the grid is closer to the cathode and the anode is relatively distant (as per G8HQP Dave's post #10), on that basis you'd rather think the '83 could cope with MORE anode dissipation than the '82, not less!
If we imagine a temperature gradient (kelvin per mm) from the hotter cathode to the cooler anode then moving the grid closer to the cathode will move it up that gradient, thereby raising its temperature. If we leave the cathode and the a-k spacing unchanged then the only way to compensate for this would be to make the gradient steeper, so the temperature falls more quickly as we move away from the cathode. Making the gradient steeper would involve making the anode cooler, not hotter.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 23rd Aug 2018, 1:07 pm   #40
terrykc
Rest in Peace
 
Join Date: Nov 2010
Location: North Hykeham, Lincolnshire, UK.
Posts: 515
Default Re: Valve construction.

Quote:
Originally Posted by kalee20 View Post
ECC82 and 83... yes I've been able to tell an unmarked one by the grid, but only if I have a pair of known ones to compare against (and that's ignoring the possibility the unmarked one is an '81).
Don't forget that, if the valves are of Philips origin, there are type codes etched into the glass. There is a definitive list on the internet here but if you have two and one is clearly identifiable, if the etched codes match, they are both the same.

Ignore the second row of etched characters - these indicate factory and date of manufacture and will obviously vary from valve to valve.
terrykc is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 3:17 pm.


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 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.