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 22nd Jan 2022, 2:52 pm   #1
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default The GZ37 - what's its history and is it a good rectifier ?

I've recently had an amp in for repair. It has a separate mains HT supply with a GZ37 in it. This supply had taken to blowing T2A fuses in the mains transformer's primary circuit at switch-on, initially only occasionally, but then more often and eventually every time and with fireworks in the GZ37.

I checked for faults outside the rectifier but couldn't find any. So I replaced it with a 'new old stock' GZ37 and all now seems well - it's run for a few tens of hours and has been switched on and off perhaps 30 times (always either with no time for the valves to warm up after switch-on, or plenty of time for them to cool down after switch-off).

The HT supply circuit is pretty simple - a 460-0-460V HT secondary with about 80ohms effective resistance in each half-winding, then the GZ37, then a 30H 125ohm choke feeding a pair of 220uF electrolytics in parallel. There's a 10k resistor from HT to ground, and the amplifier itself which draws just under 230mA quiescent. The output voltage across the resistor and feeding the amp is 312V or so.

I've modelled the supply using the Duncan Amps psud2 programme which I've always found to be very good. With essentially no adjustable input parameters (I used the separately measured properties of the mains transformer, choke and amplifier load) it predicts the steady-state output voltage very accurately (within a volt or two of what I measure) so I assume it's working well. It tells me that the maximum current in each of the rectifier diodes is less than 275mA (compared with the GZ37's 750mA rating) and the peak inverse voltage is 1.25kV (compared with the rated max of 1.85kV). So, given this relatively gentle operating regime, I'm wondering why the original valve has failed and whether I should worry about the durability of the new one ?

A quick forum search reveals this thread https://www.vintage-radio.net/forum/...d.php?t=158725 where, at post #13, PJL says "There is a similar GZ33 which tolerates a larger reservoir capacitor and it would be interesting to understand why and the intended purpose of the GZ37." I'd be interested to know too.

My first question is "When did Mullard first start selling GZ37s ?" The datasheet that I have is this one http://www.r-type.org/pdfs/gz37.pdf, dated October 1958, but I know these were sometimes revised and reissued, so does the valve pre-date this ? It's said to be an equivalent for the 53KU and that came out in 1948 or so (its release was one of the reasons that Williamson published a re-design for his famous amplifier in Wireless World, Aug 1949). It seems odd that Philips/Mullard would wait 10 years before releasing an equivalent. I do have the Mullard Valve Data Booklet for 1954 and the GZ37 is not listed there. Then again, neither is the GZ34 which was released in that year and for which data with a Dec 1952 date is included here http://www.r-type.org/pdfs/gz34.pdf.

My second question is "Isn't the GZ37 an 'old technology' valve, so why release it at all when you already have the in many ways equally good GZ34 ?". The GZ37 has larger electrodes and a more powerful heater than the GZ34, so we might expect it to be more capable. But its PIV rating is only marginally better (1.6kV/1.85kV vs 1.5kV) and its peak current rating is the same (750mA for both valves). In 'capacitor input' circuits both valves are rated up to 250mA output for output voltages up to 480V. The GZ34's current rating drops at voltages higher than this. The GZ37 datasheet doesn't give data for output voltages above 486V. One factor in that may be the high forward voltage drop in the GZ37. At 250mA output its Vf is 46V. That of the GZ34 started out (1952 data) at 27V and this was subsequently improved (1958 data) to just 16V. Incidentally, it seems quite easy to find copies of the first two pages of the GZ37 datasheet online, but I've yet to find the remaining three pages, so I've attached pictures, albeit small, of them below.

The GZ37 does have a higher output current rating for 'choke input' circuits than the GZ34 does - 350mA vs 250mA. Is that the only reason for its existence though ? If it is then it seems like only a small benefit. Were Philips/Mullard really losing a lot of business over that 100mA difference ? Or did they feel some need for their own valve, even if it was old-fashioned, once Cossor (who had developed the 53KU) finally let their valve manufacture go wholly into the hands of EMI and M-OV, Philips/Mullard's big competitor, in 1956 https://en.wikipedia.org/wiki/A.C._C...mpany_timeline ?

Cheers,

GJ
Attached Thumbnails
Click image for larger version

Name:	GZ37 Vf 8x6.jpg
Views:	70
Size:	62.3 KB
ID:	250162   Click image for larger version

Name:	GZ37 C input 8x6.jpg
Views:	79
Size:	56.7 KB
ID:	250163   Click image for larger version

Name:	GZ37 L input 8x6.jpg
Views:	67
Size:	53.6 KB
ID:	250164  
__________________
http://www.ampregen.com

Last edited by GrimJosef; 22nd Jan 2022 at 3:18 pm.
GrimJosef is online now  
Old 22nd Jan 2022, 3:14 pm   #2
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: The GZ37 - what's its history and is it a good rectifier ?

I'll start another post to include some more pictures. These relate to metalisation on the inside of the GZ37's glass at the top of the valve, both above and below the mica which is used to centre the metalwork assembly in the glass envelope.

The first picture shows the new old stock GZ37 on the right and the failed arcing one on the left. The failed one has distinct silver/dark patches corresponding to the holes in the mica. It's as if the diodes are releasing streams of metal vapour which are passing through the holes and being deposited on the glass. It's a bit hard to see, but I can say that these patches are quite heavily deposited. They're not opaque, but when front-lit the square ones at least reflect enough light that it's hard to see the mica details behind them.

The second picture shows a similar view of the new old stock valve after it's had a few tens of hours (maybe 30 or so ?) of running. The square patches are already becoming obvious on the glass, although there's no sign of metalisation corresponding to the mica slots yet.

The third and fourth pictures show side views of the valves. In the third we can see a dark band below the centring mica in the failed valve whereas the unused valve's glass is clean and clear. After a few tens of hours the fourth picture reveals that the glass has already darkened noticeably below the mica.

I confess I was surprised how quickly this metalising appeared. Do we know where it is coming from (these are indirectly heated valves, with the heater wire coated in an insulating layer, so it's not as if there is an extremely hot exposed tungsten flament to act as the source) ? Is it some residue left over after the manufacturing process is complete and which just needs to be burnt away in the first period of operation, after which the metalisation rate reduces and the darkening stops getting much worse ? Or is this relatively rapid darkening of the new valve a sign that something is driving it hard, so it may not have nearly the lifetime (a few thousand hours ?) that I'm hoping for ?

Cheers,

GJ
Attached Thumbnails
Click image for larger version

Name:	Metalised patches at top of failed GZ37 8x6.jpg
Views:	179
Size:	60.1 KB
ID:	250170   Click image for larger version

Name:	Metalised patches in GZ37 after few 10s of hours 8x6.jpg
Views:	161
Size:	43.6 KB
ID:	250171   Click image for larger version

Name:	Dark metalising in failed GZ37 8x6.jpg
Views:	149
Size:	48.2 KB
ID:	250172   Click image for larger version

Name:	Dark metalising in GZ37 after few 10s of hours 8x6.jpg
Views:	131
Size:	47.5 KB
ID:	250173  
__________________
http://www.ampregen.com

Last edited by GrimJosef; 22nd Jan 2022 at 3:23 pm.
GrimJosef is online now  
Old 22nd Jan 2022, 5:08 pm   #3
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: The GZ37 - what's its history and is it a good rectifier ?

THe slots in the mica tell you that the manufacturer was aware that there would be metal spraying around, so they wanted to lengthen the leakage path.

Were they running the heater too hard to get the last bit of emissivity?

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 22nd Jan 2022, 5:45 pm   #4
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: The GZ37 - what's its history and is it a good rectifier ?

The rectifier heater is supplied from the mains transformer in the psu, which also has the HT winding on it (all the amp's other valves run off a single separate heater transformer). The psu transformer does generate the full 5V at the GZ37 heater pins, but not significantly more. With an average current of 260mA passing through the rectifier, and about 50V Vf, the anodes are dissipating 13W or so between them. So the cathode and heater will be sitting in a relatively hot thermal radiation bath, which might raise their temperatures a bit I suppose.

As far as leakage goes, there are two micas at the top of the valve. They seem to be identical in terms of slots and holes (cheaper to make them all the same, perhaps). The upper one just centres the assembly inside the glass envelope. The lower one holds all the metalwork in place and so handles the electrical stress. But the cathode tubes poke up through this and if the metal is spraying out of the top of the tubes then perhaps most of it will not end up on the electrically-stressed lower mica. It clearly does hit the upper one, as the metalised patterns on the glass show.

Cheers,

GJ
__________________
http://www.ampregen.com

Last edited by GrimJosef; 22nd Jan 2022 at 5:54 pm.
GrimJosef is online now  
Old 22nd Jan 2022, 5:48 pm   #5
PYE 405
Pentode
 
Join Date: Dec 2009
Location: Chatteris, Cambridgeshire, UK.
Posts: 176
Default Re: The GZ37 - what's its history and is it a good rectifier ?

I have seen the same thing with a pair of NOS Mullard GZ32's in Quad II amplifiers. There is nothing wrong with the amplifiers and after some 5 years of satisfactory use, the level of deposition is still the same as it was after a few weeks when they were first installed. I believe it is is simply material being vaporised from the heaters and depositing itself on the cooler glass envelope. It happens to form in a circular pattern directly on the top glass above the cathode cylinders, the same with both valves.
I think this is normal and nothing to indicate the valve is being over driven.
__________________
Andrew
PYE 405 is offline  
Old 22nd Jan 2022, 5:58 pm   #6
PYE 405
Pentode
 
Join Date: Dec 2009
Location: Chatteris, Cambridgeshire, UK.
Posts: 176
Default Re: The GZ37 - what's its history and is it a good rectifier ?

As to the failure of the first GZ37, it could be due simply to age and back emission developing, causing conduction during the inverse portion of the AC input.
__________________
Andrew
PYE 405 is offline  
Old 22nd Jan 2022, 6:52 pm   #7
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: The GZ37 - what's its history and is it a good rectifier ?

I'd wondered about its raison d'ĂȘtre too, it seems to be a bit misplaced in the history of valve development- one possible explanation that I'd pondered was as a more robust (i.e. with its indirectly-heated cathode) substitute for the U52 family of directly-heated rectifiers. Directly-heated cathode tapes can get a bit fragile with age and U52-type applications ranging from PA kit to modest military/professional transmitters/transceivers maybe needed something more bump-proof and dependable? That could explain such seeming anachronisms as the relatively high forward voltage drop and the pinch-style construction. Power amplifier applications using high-voltage electrolytics often out of necessity push the HT voltage as high as can be reasonably got away with and maybe substituting something with forward voltage drop as low as the GZ34 would driven the HT a bit too high, and, if it was intended as a substitute for U52-type valves, then it wouldn't have needed to be as compact as the GZ34 with its tubular envelope and glass button base (within the octal lower shell), so older existing "Coke-bottle" envelope and pinch-seal plant could be re-deployed.

All a bit of shakey supposition on my part, but otherwise an odd valve to rationalise. There were lorry-loads of CV378 versions around in the past, so presumably a recognised reliable type.
turretslug is offline  
Old 22nd Jan 2022, 7:54 pm   #8
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: The GZ37 - what's its history and is it a good rectifier ?

Quote:
Originally Posted by PYE 405 View Post
... after some 5 years of satisfactory use, the level of deposition is still the same as it was after a few weeks when they were first installed ...
That's useful to know Andrew. I get to see quite a lot of valves, including unused ones on occasion, but I don't often get to spend very long watching them.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 22nd Jan 2022, 8:39 pm   #9
Guest
Guest
 
Posts: n/a
Default Re: The GZ37 - what's its history and is it a good rectifier ?

These don't have to do anything subtle like amplification or low leakage so they are run quite hard. After all they are supplying current for at least two valves the same size. Hot metal in a vacuum, bound to get some evaporation.
 
Old 22nd Jan 2022, 9:37 pm   #10
paulsherwin
Moderator
 
paulsherwin's Avatar
 
Join Date: Jun 2003
Location: Oxford, UK
Posts: 27,786
Default Re: The GZ37 - what's its history and is it a good rectifier ?

I suspect turretslug has nailed it. Mullard decided to make the GZ37 because the MOD were buying large quantities of them from other suppliers, and they wanted a slice of the pie. The GZ37 was just the civilianised version.
paulsherwin is online now  
Old 23rd Jan 2022, 12:11 am   #11
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: The GZ37 - what's its history and is it a good rectifier ?

I had tried to reverse engineer the transient peak current rating of the GZ37 using PSUD2, as it was a parameter missing from the known datasheet(s) - but the outcome was uncertain. Some (including Morgan Jones) have assumed the 'capacitor input' capability of the GZ37 was the same as the GZ34, but I strongly doubt that.

A lack of any sign of sprayed metallisation at the base of the tube is probably related to the filament being significantly cooler as it exits the cathode tube due to thermal conduction to the base headers as well as far better thermal radiation to free space. In comparison, the filament at the top end of the tube has a hotter nearby cathode tube and only a small spherical angle for thermal radiation to free space, and the U-bend in the filament may be exposing filament wire through cracks in the filament insulation coating.

FWIW, I'd recommend adding series 1N4007's to the GZ37 anodes for just that type of end-of-life scenario. Not only do the ss diodes avoid collateral damage to the power transformer, but they can extend the economic service life of a GZ37, given the price of NOS.
trobbins is offline  
Old 23rd Jan 2022, 5:11 am   #12
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: The GZ37 - what's its history and is it a good rectifier ?

I'd avoid the high-voltage end of the 1N400x family. They are really PIN diodes and turn off with a snap, creating rather a lot of RF energy and radio interference. They are barely fast enough for use on 50Hz supplies.

Counter-intuitively, ultra-fast rectifier diodes are a lt cleaner, and generate less stress.

Added to a valve rectifier, the valve turns off first and saves the day, but the stored energy in the 1N4007 still has to go somewhere.

As an alternative, you could use solid state rectifiers, then followed by the GZ37 as a DC voltage dropper and adding softness to the effective supply impedance. This way you can strap both anodes together and you no longer have the fireworks mode of anode-to-anode arcing. As a bonus, electricity can be generated from purists spinning in their graves

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 23rd Jan 2022, 5:49 am   #13
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: The GZ37 - what's its history and is it a good rectifier ?

A 1N4007 in series with a GZ37 anode turns off like a GZ37 anode - no RF or snap or stored energy (due to the long and slow off commutation event). It just takes a bit of effort to measure the dI/dt in that type of circuit and relate that to the dI/dt it takes to force an ss diode in to reverse recovery, and to how slowly the current falls to zero and hence how negligible any 'stored charge' is.

I tried to identify any disturbance last year using a worst-case choke input full-wave B+ supply with significant dc loading and using either 1N4007 and UF4007 diodes. My soundcard (out to 96kHz) could just make out some resonant content on the HT secondary winding about 100dB below the mains fundamental that aligned with the PT HT half-secondary inductance ringing with its local shunt capacitance, and that was down near the noise floor - so far below what a scope could discern. I measured the PT winding leakage inductance and shunt capacitance to confirm it aligned, and added shunt capacitance to confirm the resonance frequency dropped in kind, and added a tuned HT winding snubber (ie. quasimodo style) to confirm that suppressed any related signal.

Last edited by trobbins; 23rd Jan 2022 at 6:06 am.
trobbins is offline  
Old 23rd Jan 2022, 9:06 pm   #14
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: The GZ37 - what's its history and is it a good rectifier ?

Thanks for this. It's all been helpful information. It's left me less concerned about the rectifier in my customer's psu than I was at the start.

Checking the datasheet for the CV378 http://www.r-type.org/pdfs/cv378.pdf I see that it's dated 1954 and that the specs are slightly different from the GZ37 (e.g. the CV378's max working PIV is just 1.25kV but its peak anode current is 1.25A). The wide range of envelope diameters will accommodate both the thinner, later shouldered variety and earlier, fatter 'Coke bottle' type.

Looking at the picture of the Cossor 53KU/GZ37 here http://www.r-type.org/exhib/aaa1092.htm it's interesting that Cossor (not a Philips/Mullard company) have chosen to print their competitor's type number for this valve alongside their own. Given that Mullard seem to have been a late entrant into the market I'd have thought Cossor would just have ignored them.

One last question - does anyone have any Mullard publicity or technical material from before 1958 which mentions the GZ37, or which doesn't when you might expect it would if the valve existed ?

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 24th Jan 2022, 10:55 am   #15
woodchips
Octode
 
Join Date: May 2010
Location: Grantham, Lincolnshire, UK.
Posts: 1,172
Default Re: The GZ37 - what's its history and is it a good rectifier ?

Isn't there a problem with the two of 220uF smoothing caps?

I only have a WW valve data book, no maximum smoothing capacitance stated, but for most other similar rectifiers it is only 50uF or so? The U54 and 53KU is just 16uF.

Been quite some time since I read it, but the article Analysis of Rectifier Operation in the RCA Electron Tubes book had, from memory, some pertinent information.
woodchips is offline  
Old 24th Jan 2022, 3:05 pm   #16
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: The GZ37 - what's its history and is it a good rectifier ?

If the HT filter was 'capacitor input' then yes, 440uF would be way too large a cap for the first element (commonly called the reservoir cap). But this supply is 'choke input' i.e. the first element after the rectifier is a great big choke. This limits the rate of change (particularly the rate of rise) of current so effectively that almost whatever capacitance follows it won't present a problem for the rectifier.

The psud2 model can simulate what happens at switch-on but it has to assume that the rectifier comes to life instantly. Even in those circumstances the peak current is still less than 800mA, and it's that high only for a relatively small number of mains cycles. Given that the rectifier actually warms up quite slowly the peak current in real life is likely to be less still.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Old 25th Jan 2022, 12:47 am   #17
trobbins
Heptode
 
trobbins's Avatar
 
Join Date: Oct 2009
Location: Melbourne Australia
Posts: 898
Default Re: The GZ37 - what's its history and is it a good rectifier ?

The hassle with the GZ37 is that Mullard and Chelmer datasheets show a 4uF max capacitor input filter value. PSUD2 shows the continuous peak current is 0.71A with 485V output for 250mA load with a transformer secondary of 500V, which aligns with the datasheet. Simulating the choke related parameters in PSUD2 is not well aligned, with the output voltage result being noticeably low.

For the 4uF capacitor input filter benchmark, PSUD2 shows that the continuous peak current falls for lower transformer secondary voltages, and that a hot turn-on event at 500V would cause an initial peak current of 0.93A for worst-case 60Hz mains.

Given that peak currents fall with lower transformer voltage, one could use PSUD2 to appreciate that with a 300V HT, and using the same 250mA DC load limit and PT secondary resistance, the filter cap could be increased only to 10uF before the hot turn on current exceeds a 0.93Apk.

Even with a choke input filter, there are limits to how much capacitance can then be used. With the datasheet 500V example conditions for choke input, the output capacitance is limited to about 150uF before a hot turn-on peak exceeds 0.93A - not that 150uF was envisaged as a B+ capacitance back then.

Exceeding such a transient peak current limit is going in to unchartered territory with respect to service life and cathode degradation from single events.
trobbins is offline  
Old 25th Jan 2022, 2:11 am   #18
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: The GZ37 - what's its history and is it a good rectifier ?

The GZ37 was listed in Radio Valve Guide Book 4, which covered releases in the period 1956 to 1960. So that is consistent with the putative 1958 release date.

On the other hand, the Cossor 53KU was listed in Book 1, covering releases in the 1934-51 period.

Although Cossor had its own valve designation system, in the post-WWII era it appeared to have been moving away from this, largely using the American system. E.g., it did the American registration of the Z77 as the 6AM6. (And presumably Philips/Mullard did the Pro-Electron registration as the EF91.) Whilst Cossor may have preferred the relatively random American system, possibly it was not averse to using the more systematic and logical Pro-Electron system where appropriate, simply for the benefits of having a standard classification code on the valve, which in this case became available once Mullard adopted and registered the 53KU (or something very close to it) as the GZ37.


Cheers,
Synchrodyne is offline  
Old 25th Jan 2022, 9:43 am   #19
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,799
Default Re: The GZ37 - what's its history and is it a good rectifier ?

Although the choke-input filter following a full-wave rectifier helps a lot with peakcurrents, there is a price.

THey can operate in either of two modes. If the choke is big enough, or if the minimum current is high enough, they carry continuous current. Discharge of the choke makes up for when the transformer voltage is lower than the filter capacitor voltage. So therectifier diodes change conduction immediately from one to the other. they don't care whether the energy comes out of the transformer or the choke. In this mode, the capacitor sits at close to the mean of the modulus of the transformer voltage.

If the load current is too little (for the mains frequency and the choke inductance) conduction will cease twice per cycle as the choke runs out of oomph. In this discontinuous mode, the capacitors will now charge to the peak, not the mean. So off load, the output DC shoots up.

Choke filters work well in class-A audio amps, there there is always plenty of current demand. If you have to do one for an application where the current varies a lot, then you need high inductance to handle the low current situation, and you need high saturation current to handle the high current case... so your choke winds up being huge.

Sometimes you just can't win.

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 25th Jan 2022, 10:11 am   #20
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: The GZ37 - what's its history and is it a good rectifier ?

The standard solution amongst audio designers, at least more recent ones, to the problem of keeping the output DC down seems to be simply to load the supply with a resistor hard-wired across its output. In the case I've been dealing with it's a 10k Al-clad wirewound heat-sunk to the case. The price which is indeed paid is 10W of waste heat.

The choke lamination stack still has external dimensions of 4.5" x 3.75" though, and is 1.75" thick. You wouldn't want to drop it on your foot.

Quote:
Originally Posted by trobbins View Post
... Even with a choke input filter, there are limits to how much capacitance can then be used. With the datasheet 500V example conditions for choke input, the output capacitance is limited to about 150uF before a hot turn-on peak exceeds 0.93A - not that 150uF was envisaged as a B+ capacitance back then.

Exceeding such a transient peak current limit is going in to unchartered territory with respect to service life and cathode degradation from single events.
I feared that quite a few hours of work would be needed to get a grip of what really happens at switch-on. A capacitor as large as 440uF has a discharge time of the order of a second, during which the rectifier cathodes will cool somewhat. So a true hot switch-on, into an uncharged capacitor, is hard to achieve. As the cathodes cool their emissivity drops, of course, so peak currents may become limited by what the cathodes can deliver. We might think this is a safety feature, keeping the current lower than it might otherwise be. But in fact if the cathode can't supply what the load is trying to draw then there will be some degree of stripping of the protective space-charge layer around it and the risk of damage as a consequence.

When my customer went home with the amp yesterday I did recommend that he wait at least a minute, but ideally two, between switching the amp off and switching it back on again.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is online now  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 4:04 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.