Amps
[Top] [All Lists]

Re: [Amps] Ameritron (et. al.) filament voltages

To: Rob Stampfli <rob@cboh.org>
Subject: Re: [Amps] Ameritron (et. al.) filament voltages
From: Roger <sub1@rogerhalstead.com>
Date: Fri, 29 Jan 2010 01:13:38 -0500
List-post: <amps@contesting.com">mailto:amps@contesting.com>

Rob Stampfli wrote:
> Hi Alek,
> Thanks for the response.  Personally, I agree with you that it is better
> to keep the filament voltages in check for a number of reasons.  However,
> here's a pointer to a post (from this list) that takes the opposite
> perspective:
>
> http://lists.contesting.com/archives//html/Amps/2007-07/msg00367.html
>
> Tom is Tom Rauch, who had a hand in the design the AL-80B, and who
> certainly knows more about building amps than I ever will.  Note he's
> not disputing the Eimac claim, only counter-claiming that the hazards of
> ICAS operation will typically kill the tube before an over hot filament
> will.  He may well be right, but I still can't see any advantage to
> running excessive voltages on the filament.  That's why I cut mine back.
>
> I put the question to the list in hopes that someone could explain why
> commercial amps always seem to be designed for higher filament voltages.
> The only advantage I can see is that it gives the amp more headroom in
> the event of low line voltage or soft line voltage that falls excessively
> under load.  Does anyone know of any additional advantages?
>   
I think it gets back to line voltage regulation.  I say that as I'm 
heavily involved with the old boat anchors and most of those guys will 
go to extremes to make sure line voltages are down to what they were in 
the "old days".  In particular, with transmitters and receivers it's a 
heat issue. Insulation and dielectrics dry out and deteriorate. 
Transformers overheat and live short lives.  Although solid state 
rectifiers reduce the heat they increase the plate voltages  and are 
often avoided.

I think part of it is to allow for low line voltage and poor regulation, 
but many of the higher priced amps have transformer taps with 
instructions on where to set them with specific line voltages.  This is 
particularly important with today's solid state, high power amps.
Many amps have the taps and instructions on how to set them for 
particular line voltages, but with the exception of those solid state 
amps very few "that I've seen" emphasize the importance of keeping the 
voltage proper.

Tom is quite knowledgeable, but there are a lot of variables when it 
comes to tube failure. There are many tubes out there than far more 
hours on them than a ham would put on them in a lifetime. I don't know 
if he's right or not as far as failures of use time Vs elevated filament 
voltage. "I believe" that tubes operated the same and treated well 
through their life will reach the point where statistically Eimac is 
correct, but when normal use is taken into consideration Tom is likely 
correct. I base my conclusion on study, but not enough evidence from a 
controlled environment to be more than anecdotal. So my opinion is just 
that, albeing what might be called an educated opinion, but remember 
that is still an opinion. At least is wasn't just pulled out of the 
air...or elsewhere. <:-))

Some specific examples:
   Remember most many of the more modern amps use indirectly heated 
cathodes. 8877, FU728, 4CX1500B 8873, 74 and 75, and even the high 
powered GU-84B/4CX-2500, 3 and 4CX-800Bs and the GS-36B/4CX400B and a 
lot of others. But take for instance the 4CX1500B Vs the 4CX1500A with 
the latter being thoriated tungsten.  running 5V and 38.5 amps +/-2A  
which is +/- 5% and 192 Watts. The 4CX-1500 B is an indirectly heated, 
oxide coated filament running 6V @ 10A +/- 10% or 60 watts which is one 
whale of a difference.  That's 132 watts that is pure heat inside the 
amp. The FU728B is 8.5-9A (5%) at 9V =~81 watts   The 3CX-2500 runs 7.5 
volts @ 51.5Amps is 386 watts.(no tolerance given)   The 4CX-3000 is 9 
Volts at 41.5 Amps = 373.5 watts(no tolerance given)

I give these examples as showing the tube filament tolerances for 
thoriated tungsten filaments is normally less than the line voltage 
tolerances and that these filaments use far more power than the typical 
indirectly heated cathode. They also appear to be more sensitive to 
voltage excursions, BUT that is based on manufactures specifications and 
those vary between manufacturers for the same tube.

So, based on all this information, I think both Tom and Eimac are 
correct for the conditions specified. The Eimac specs were quite likely 
based on continuous service at specific conditions and here the results 
of higher filament voltage could be shown, but given real world 
applications, with lots of start/stop cycles, various tune ups, not all 
of which conform to a set procedure, over driving, under loading, plus 
many other abuses, I'd have to agree with Tom that the way the tube is 
treated at least in Amateur service, probably will cause failure before 
the excess filament voltage would (within reason)

Now, having said all that, giving all those reasons, I'm going to try 
and treat those expensive (to me) tubes properly and keep the filament 
voltage where it belongs as well. I'm also going to use a soft start, 
but it stepped relay, or a slowly ramped up filament voltage.  Most 
tubes used in amateur service use indirectly heated cathodes, but the 
higher powers are usually Thoriated tungsten when may easily draw well 
over 300 watts.  The problem with this is those line variations.  If you 
don't have some method of monitoring the line voltage continuously the 
tube could easily exceed it's specified voltage many times over its 
life. In some areas of the Us and world these variations are common and 
occur often. In other areas they are few and sometimes very far between. 
Here the voltage seems quite stable, but other areas where I've lived 
(within 50 miles) the voltage has been very unstable. Here it's just 
inconsistent with lots of failures.  I run a 9500 watt generator with a 
transfer switch.  Since the spring of 2000 I've put well over 100 hours 
on that generator.  The power lines come to this rural subdivision 
through a mile of woods.  Due to austerity programs they power company 
has not kept up with the tree trimming.  The area is rather swampy with 
lots of big trees, but the trees tend to be fast growing with poor root 
systems.  They grow fast and large, but the wood isn't good for much.  
Within a year after the trees die, they become "structurally unsound" 
and it only takes a wind, or wind with ice and down they come, often 
across the power line. As the power poles carry both a ground and the 
cable system a tree falling across them shuts everything down. So even 
though the voltage is relatively steady, we get a lot of failures with 3 
or 4 per year.

I don't know if any one will find the above information and opinions 
useful, but I hope they are useful and/or informative to at least a few.
 
There must be some engineers who read this reflector who can tell us 
what the specifications are for voltage regulation in volts and %.

73

Roger (K8RI)
> Inquiring minds...
> Rob / KD8WK
>
> On Fri, Jan 29, 2010 at 06:05:43AM +0800, Alek Petkovic wrote:
>   
>> G'day Rob
>>
>> It is not any "camp" that makes the claim of shortened emissive life due 
>> to high filament voltage. It is Eimac, in its "Care And Feeding" book. 
>> For every 3% increased in filament voltage, the life is shortened by 50%. 
>> In the case of Ameritron's 10%, one would expect around 12% of normal 
>> tube life. (Didn't someone recommend Ameritron here a couple of days ago 
>> on a different thread? Ha ha)
>>
>> Yaesu have the same problem with the 2100 series amps. Even when the  
>> mains input voltage is reduced to the specified 230 Volts, the filament 
>> voltage still needs to come down.
>>
>> Like you, I have made the correction with every amp that has ever had  
>> the lid removed in this house.
>>
>> 73, Alek
>> VK6APK
>>     
> _______________________________________________
> Amps mailing list
> Amps@contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>   
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>