[Amps] Miller-Larson effect on thoriated tungsten filaments / neat electronic soft-start

[Leigh Turner]

Yes those GE CL-60 NTC inrush limiting thermistors are helpful albeit still
yield a fairly fast rise time constant in the filament voltage.  The far
gentler slow 3 to 4 second ramp up from the Triac controller is more
elegant. The slow soft-start circuit is based on the ON Semiconductor
TDA1085C IC chip and a 40A Triac.

 

In fact I've also got a couple of CL-60 thermistors (one in each leg) the
primary of the Kenwood TL922 amp's filament transformer to trim / fine-tune
the steady-state filament voltage measured right at the tube socket to circa
4.9 Volts, as the factory stock filament voltage is on the high side; around
5.2 to 5.3 Volts. That slight over voltage can also shorten the 3-500Z tube
emission life.

 

Leigh

VK5KLT

 

  _____  

From: Mike Waters [mailto:mikewate at gmail.com] 
Sent: Tuesday, 25 June 2013 3:34 PM
To: Leigh Turner
Cc: amps at contesting.com
Subject: Re: [Amps] Miller-Larson effect on thoriated tungsten filaments

 

Thanks, Leigh. If nothing else, those books emphasize the importance of a
soft-start circuit.

Evidently, it's the change in temperature that affects the life. If holding
the temp constant at those critical temperatures would actually rapidly
destroy the filament, I can't say.

A good friend of mine, WW8N, uses CL-60 current limiters in the 240 vac
PRIMARY circuit of the filament xfmr of his 3-500Z amp. These limiters are
10 ohms cold and a fraction of an ohm at rated current. 

http://www.rfparts.com/inrush.html

73, Mike
www.w0btu.com

 

On Tue, Jun 25, 2013 at 12:48 AM, Leigh Turner
<invertech at frontierisp.net.au> wrote:


Yes indeed Mike, that's the reality with these tubes.

I like to think we can mitigate at least some of this tube life degradation
through soft-start circuitry. My favourite means that has great merit is an
automated slow-ramp up of the filament voltage over a time period of several
seconds implemented with a phase-controlled Triac on the AC primary of the
filament transformer.

I have such electronic slow 4 second soft-start circuits implemented on my
twin 3-500Z tube based TL-922 amp, as well as my other ceramic-metal tetrode
based amps that run oxide-coated heaters.

Now whether it's just a pointless feel-good measure or actually yields a
practical benefit in extending filament/heater service life I cannot say.

However, gently controlling cold filament inrush current in this way
certainly seems a sensible thing to do.

Of course the plate HV xfmr also benefits from the slow ramp-up in AC mains
to prevent the large turn-on current surge of the initially uncharged HV
electrolytic capacitors over taxing the amp's ON/OFF switch contacts.

Leigh
VK5KLT