|
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. |
|
Thread Tools |
22nd Mar 2018, 7:41 pm | #1 |
Hexode
Join Date: Apr 2008
Location: Ramsgate, Kent, UK.
Posts: 252
|
Valve heaters on constant current (CCS)
Has anyone any experience of running a single heater or series heater chain from a constant current source?
I have been testing this idea using an LM317 regulator strapped as a 300mA current source in series with a 14V DC supply and a B9A socket heater pins, but the results have been a little unexpected, maybe I am overlooking something here. If I insert an EF80 into the socket the voltage across the heater pins rises to around 6.3V DC so far so good. If I swap the EF80 for a 30F5 (same pin-out) the voltage rises to around 6.5V when the manufacturers rated voltage for the valve is 7.4V @ 300mA (about 12% low). The current through the filament in both cases is exactly 300mA. This same result occurs for a handful of EF80's and 30F5's that have been tested. Also the current source is not oscillating at any time. Can anyone see the reason for the apparent discrepancy in voltage readings, I am assuming that valve heaters are (purely) resistive whether operated at DC or 50Hz AC, so ohms law should apply? Is there something, perhaps, in the way that manufacturers rated filaments for series/parallel operation or even AC/DC operation? Maybe parameter tolerances are wider than I thought? Adrian. |
22nd Mar 2018, 8:08 pm | #2 |
Dekatron
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
|
Re: Valve heaters on constant current (CCS)
Rather a long time ago I tried to work my way through the forest of claims about ECC88 and PCC88 heaters. The details are in post #16 here https://www.vintage-radio.net/forum/...ad.php?t=51821. Exec summary: Different manufacturers labelled different valves with the same type numbers (or the same valves with different type numbers). If we really care about heater characteristics at the 10% level then we probably have to measure them ourselves.
Cheers, GJ
__________________
http://www.ampregen.com |
22nd Mar 2018, 8:37 pm | #3 |
Hexode
Join Date: Feb 2014
Location: Southampton, Hampshire, UK.
Posts: 419
|
Re: Valve heaters on constant current (CCS)
Just thinking here, have you tried plotting voltage against current? That is % voltage change for a % current change at the +/- 300 mA level. With a chain there is a type of self limiting that you won't have with a fixed 300 mA.
Pete |
22nd Mar 2018, 8:44 pm | #4 |
Guest
Posts: n/a
|
Re: Valve heaters on constant current (CCS)
A radio with a dropper resistor approximates a constant current source, they seem to work OK.
|
22nd Mar 2018, 10:04 pm | #5 |
Rest in Peace
Join Date: Sep 2008
Location: Solihull, West Midlands, UK.
Posts: 4,872
|
Re: Valve heaters on constant current (CCS)
The 30F5 (300mA heater) is also known as PF818 and 7ED7. There is also 6F23/EF812/6EL7 - very similar characteristics but nominally 6.3V heater. I have seen both of these valves with the same Brimar factory code: 1291. As a general rule, a 300mA valve might not generate exactly the expected voltage drop while a 6.3V valve might not draw exactly the expected current - but in both cases the valve should work OK.
|
23rd Mar 2018, 11:53 am | #6 |
No Longer a Member
Join Date: Oct 2016
Location: Maroochydore, Queensland, Australia.
Posts: 2,679
|
Re: Valve heaters on constant current (CCS)
One advantage of using a constant current source, over a voltage source for a valve heater or filament, is that it avoids initial surge currents when the heater is cold.
If valve heaters are run from a voltage source with a low internal resistance (such as the mains power supply or a lead acid battery or a supply with a low internal resistance) and the heater is cold, the surge currents can fuse the heater. This is dealt with with Brimistors etc in series heater chains on mains power, or sometimes adding a series resistor. If it is happening you see a nasty flash from the heater at turn on. However there are some interesting scenarios/traps. In a typical small valve radio, due to the physical size of the power transformer, at turn on, with cold heaters, the total heater load is initially enough to transiently crash the heater voltage to a low level, so all the heaters get a soft start. In large chassis multi-valve sets such as TV's and amplifiers, the same usually happens because the power transformer's power rating matches the total running heater load. However, in a large chassis multi valve set, if all the high power and medium heater power output valves are unplugged, leaving just one or two small valves in circuit, the hefty power transformer (without the other large heaters loading it) is able to apply the heater potential immediately without its output sagging down at turn on and that can cause damage and even fuse the heater in a cold small signal valve. I once saw this happen in a TV tuner with all the other tubes removed from the TV's chassis in a mains transformer set. Constant current drive would eliminate the above issues, but I would be cautious of one thing. If there is an electronic failure in the circuitry the voltage on the heaters could ramp up too high. So I would recommend a series fuse and TVS or power zener devices say at least in the 7 to 8V region for 6.3V heaters, just in case. A while back I built a regulated power supply and it supplied 5V to vintage 201A valves from a 12V source, so I added TVS/zener devices & a fuse to protect those filaments in case of a power supply anomaly/failure. |
23rd Mar 2018, 12:32 pm | #7 |
Dekatron
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,951
|
Re: Valve heaters on constant current (CCS)
Apart from big epensive professional-type transmitting valves (where the magnetic field caused by high inrush current in the filament can cause distortion of the electrode assembly) I can't see the need for constant current heaters.
Old TVs with series-connected heaters wired across the mains - several different styles of valves in use - all with different warmup characteristics - I well recall seeing some of the smaller lower-thermal-inertia valves [EB91 diodes!] in a series-chain glowing bright as light-bulbs for a few seconds at cold-switch-on, and they never seemed to suffer any undue harm. [I've heard it said that the only way to kill an EB91 for sure is to hit it with a shovel]. Of the valves I've had to replace, only a very small proportion have been due to heater failure: they're more likely to go 'soft', or low-emission. |
23rd Mar 2018, 12:53 pm | #8 |
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,184
|
Re: Valve heaters on constant current (CCS)
According to the Philips data books, when the valves are specified for series heater use, current is the most important parameter and should be kept within 5% for best results. Tension is only a secundary characteristic. For parallel heaters, this rule is the other way around.
|
23rd Mar 2018, 1:31 pm | #9 |
Hexode
Join Date: Apr 2008
Location: Ramsgate, Kent, UK.
Posts: 252
|
Re: Valve heaters on constant current (CCS)
Thanks for the replies chaps, they have set me thinking. I believe that GrimJosef & G8HQP Dave have hit the nail on the head, that the tolerances on heater V/A are actually as wide as a barn door, regardless of manufacturer's data. After spending my entire career designing high reliability electronic kit to tight specifications it is difficult to switch mindset!
The reason for trying this out is that I have set myself the task of creating a reasonably good quality glass based audio amplifier out of readily available and sensibly priced parts, so from the glassware point of view that really means using TV type valves. This suggests series heaters because of the different voltage ratings, the use of a CCS being a nice way of allowing an off-the-shelf transformer with a standard voltage rating to match the total chain voltage required. The added bonus (as Argus25 says) is that the normal inrush current of the filaments is held off, although it takes 40 seconds+ for the voltage to ramp up to its operating value. Other bonuses are that using DC will make the amp inherently less 'hummy', and the ability to swap valves like EF80/30F5 and ECF80/PCF80/30C1 is made possible (humour me on this one!). And, yes, I know that valves like the EF80 make good microphones, that is another thing to solve. I will post in the Homebrew Section of the forum when there are some physical results of my efforts to show. Adrian |
23rd Mar 2018, 1:42 pm | #10 |
Hexode
Join Date: Apr 2008
Location: Ramsgate, Kent, UK.
Posts: 252
|
Re: Valve heaters on constant current (CCS)
Just crossed my last post with a couple of others.
Thanks Maarten, that makes a lot of sense. G6Tanuki, yes, I do remember as a teenager having harvested a large crop of EF80 valves from discarded TV sets (you could roam the rubbish tips and take what you wanted in those days!). One day, I must have been bored, and connected the heaters to a variable voltage PSU in the school physics lab but did not manage to destroy any although they glowed like lightbulbs! Adrian. |
23rd Mar 2018, 3:38 pm | #11 | ||
Dekatron
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,059
|
Re: Valve heaters on constant current (CCS)
Quote:
Quote:
I've found the later-generation valves remarkably predictable as far as V/I is concerned! Switch-on surge is a point well made. Series strings are fairly well protected, as long as there ar not bucketloads of valves such that the voltages add up to nearly the full (stiff!) mains voltage. Parallel connection from a transformer less so, unless the transformer is inefficient and lossy, in which case the 'sag' in voltage with the low-resistance cold load gives them an easy life. |
||
23rd Mar 2018, 5:59 pm | #12 |
Dekatron
Join Date: Aug 2003
Location: near Reading (and sometimes Torquay)
Posts: 3,086
|
Re: Valve heaters on constant current (CCS)
I would have expected in the cases where a valve might be labelled P or E because it is both, that the P variant probably has better H-K insulation thus making it more suitable for series chain use.
|
23rd Mar 2018, 9:12 pm | #13 | |
Heptode
Join Date: Mar 2011
Location: Dorset, UK.
Posts: 519
|
Re: Valve heaters on constant current (CCS)
Quote:
However the same could not be said for the Mullard EH90 under the same conditions. This one stayed at just below 5.9V! Possible since this valve was designed for series connected operation. A similar situation with PL81 where the heater voltage finished at 21.25V
__________________
To an optimist a glass is half full; a pessimist half empty; an engineer twice as big as need be! |
|
23rd Mar 2018, 10:06 pm | #14 |
Dekatron
Join Date: Aug 2013
Location: Wigan, Greater Manchester, UK.
Posts: 9,427
|
Re: Valve heaters on constant current (CCS)
From around 1963, Mullard valves for series heater chains had a controlled warm up.
http://www.r-type.org/addtext/add126.htm
__________________
Frank |
23rd Mar 2018, 10:22 pm | #15 | ||
Dekatron
Join Date: Dec 2007
Location: Haarlem, Netherlands
Posts: 4,184
|
Re: Valve heaters on constant current (CCS)
Quote:
|
||
24th Mar 2018, 1:30 am | #16 |
Dekatron
Join Date: Jan 2012
Location: Brentwood, Essex, UK.
Posts: 5,316
|
Re: Valve heaters on constant current (CCS)
The GEC-Osram valve catalogue for 1938-9 mentions that its octal valves includes a "Universal" range having 6.3V, 0.3A heaters that were designed for either series or parallel operation. The relevant individual data sheets give the heater voltage as 6.3V, and the current as 0.3A (approx).
Last edited by emeritus; 24th Mar 2018 at 1:36 am. |