[Amps] plate dissipation and duty factor

Roger sub1 at rogerhalstead.com
Thu Mar 12 20:17:41 PDT 2009



Jeff Blaine wrote:
> Gentlemen,
>
> I was thinking about this anode disipation capability today as well.  
> The ability to pull the heat off the plates is going to be limited by 
> the thermal gradient between the anode face and the disipation 
> structure.  And I guess that is a function of the thermal resistance 
> between the two structures. 
>
>  From Rogers comment, I was thinking (in the extrememe SSB case) about 
> voice peaks which are transient by nature.  Depends on voice 
> characteristics.  In any event, the longer the current flow exists at 
> these higher levels, the rate of heating of the anode is also maximal.  
> So at some finite pulse width, the point of damage would be reached even 
> if a guy were thinking "I'm in a low duty cycle mode."
>
> That makes the scaling below theoritically possible, but not practical.  
> So further derating or bounding of the anode dis limits is called for.  
> Clearly the cathode needs to be able to "give" it just as the anode 
> needs to be able to "take" it.
>   
How about water cooling?  That will greatly increase the plate 
dissipation capability, but how far can you go before cathode emission 
becomes the limiting factor, or secondary emission?

73

Roger (K8RI)
> For the cathode emission limts, do you think there is a way to 
> characterize this?  Or perhaps infer it from pulse-mode data assuming 
> comprable CCW data is not available?
>
> 73/jeff/ac0c
>
> Karl-Arne Markström wrote:
>   
>> The assumption that duty cycle * peak dissipation =  average 
>> dissipation holds so long as the pulse lengths or envelope variations 
>> are short compared to the thermal time constant of the anode structure.
>>
>> For small tubes, the time constant is in the order of a few seconds, 
>> and is in the order of 30 - 60 seconds for large tubes.
>>
>> Sometimes, the time constant can be deduced from the
>> tube data sheets, an example is the Telefunken RS2795 (~4CX15000A in 
>> European disguise) where the continuous plate dissipation is given as 
>> 25 kW, and the permissible dissipation for 30 seconds as 35 kW.
>>
>> As K5PRO stated, the maximum available power within a distorsion 
>> specification is more often than not limited by the peak cathode 
>> emission.
>>
>> 73/
>>
>> Karl-Arne
>> SM0AOM
>>
>> ----- Original Message ----- From: "Jeff Blaine" 
>> <keepwalking188 at yahoo.com>
>> To: <jtml at vla.com>
>> Cc: <amps at contesting.com>
>> Sent: Thursday, March 12, 2009 8:16 AM
>> Subject: Re: [Amps] plate dissipation and duty factor
>>
>>
>>     
>>> Hi John,
>>>
>>> Thanks very much for the info.  That's a great bit of data there.  And
>>> it's great to have that confirmation.
>>>
>>> For the duty cycle, the ARRL if I remember correctly calls out the duty
>>> cycles for unprocessed SSB at 20%, and processed SSB as well as
>>> "conversational" CW both at 40%.  And I want to say that PSK runs about
>>> 70% - something like that.
>>>
>>> So to combine all this together, let's say that we have a CCS plate dis
>>> spec of 100W.  And assuming the scaling effect (duty cycle vs plate dis)
>>> is completely linear, then that means we should be able to safely run at
>>> these power levels - assuming here that we are only talking about the
>>> tube to simplify the discussion:
>>>
>>> Mode                     duty cycle      plate dis
>>> --------------------------------------------
>>> RTTY                        100%          100w
>>> PSK (approx)              70%         140w
>>> Unprocessed SSB        20%         500w
>>> Processed SSB/CW      40%         250w
>>>
>>> What do you think?
>>>
>>> 73/jeff/ac0c
>>>
>>>
>>>
>>> John Lyles wrote:
>>>       
>>>> You are correct that RTTY is basically 100% on like FM, at least 
>>>> during a burst of transmission. So this is limited by the average 
>>>> plate dissipation of the tube, not to exceed XXX watts or KW. 
>>>> Thermally, a few seconds on is like CW as far as the tube anode 
>>>> structure is concerned.
>>>>
>>>> For pulsed ratings, the average dissipation limit is the same. If 
>>>> you are on 10% of the time, then the peak power may go up to the 
>>>> limits of the cathode emission current of the tube in some cases, 
>>>> however, the plate dissipation still sets the limit for the thermal 
>>>> loading on the anode. If you run 10 X the CW power level and are on 
>>>> 10% of the time, its the same dissipation as CW. You cannot cheat 
>>>> and get more dissipation on an average than the tube is capable of. 
>>>> You can get high peak dissipation, but that really isn't the 
>>>> definition of dissipation. Its an average thing. There is a limit to 
>>>> how long you can leave the pulse on, however, before the tube makers 
>>>> say no. In other words, you cannot run, like 1 second on at 10 X the 
>>>> average dissipation, then wait 10 seconds and do it again, and 
>>>> expect long life. Fusion RF systems run a blast of RF for 20-100 
>>>> seconds, and they are considered CW. Particle accelerators like 
>>>> where I work run 10% DF, so we can crank up the peak to quite
>>>>         
>>>  h
>>>       
>>>>  igh. We
>>>> have 250 kW of plate dissipation in triodes, and on a peak basis its 
>>>> like 2.5 MW. But thermally it is only 250 kW of power into the 
>>>> copper. The pulses are short.
>>>> Oxide cathode and thoriated tungsten tubes have their favored 
>>>> regimes with respect to peak cathode emission, which is often 
>>>> another unmentioned limit of a tube.
>>>>
>>>> As for SSB, I defer to those who practice making SSB amplifiers.
>>>>
>>>> 73
>>>> John
>>>> K5PRO
>>>>
>>>>
>>>>         
>>>>> Message: 7
>>>>> Date: Wed, 11 Mar 2009 21:51:34 -0500
>>>>> From: Jeff Blaine <keepwalking188 at yahoo.com>
>>>>> Subject: [Amps] What is the true and actual meaning of a plate
>>>>> dissipation limit?
>>>>>
>>>>> Gentlemen,
>>>>>
>>>>> Wanted:  an understanding of the actual and true meaning of the plate
>>>>> disipation limits with respect to duty cycle.
>>>>>
>>>>> I may have missed it, but it seems the answer is hiding at least 
>>>>> from me.
>>>>>
>>>>> All our tubes have plate dis limits associated with them.  Sometimes
>>>>> there is an associated cooling requirement with it as a footnote, but
>>>>> beyond that, not much else is said.
>>>>>
>>>>> Say a guy loves SSB (low duty cycle) and RTTY (100% duty cycle).  The
>>>>> rule of thumb in some cases is to run the RTTY mode at 1/2 the typical
>>>>> power of SSB.  But this is often stated without explaining why the 
>>>>> RTTY
>>>>> power level specified as 1/2 is the right level from a 
>>>>> specification or
>>>>> design standpoint.
>>>>>
>>>>> I realize that in the greater scheme, there are a host of 
>>>>> components to
>>>>> consider when talking about an amp as a whole.  But here i am 
>>>>> addressing
>>>>> the tube only as an isolated case.
>>>>>
>>>>> Eimac's C&F does not mention RTTY that I recll, but they do talk a lot
>>>>> about commercial 24/7 FM service - and that's a 100% non-stop mode;
>>>>> equivalent to RTTY.  They suggest in the C&F documents that the tube
>>>>> will run up to the rated plate dis in CCS.  OK.  Maybe the
>>>>> interpretation is that the Pd-max is a hard limit?  Valid for all time
>>>>> and all cases.
>>>>>
>>>>> And then there are the pulse applications that come along and spoil 
>>>>> the
>>>>> CCS argument.  Many tubes have a pulse rating - or in the case of many
>>>>> of the Russian tubes - a pulse rating spec set only without CCS duty
>>>>> being adequately specified.
>>>>>
>>>>> In these pulse duty cases, the time averaged plate dis is below the
>>>>> published limit, I'm sure.  But for the pulse duration, the Pd is 
>>>>> going
>>>>> to be exceeded by a huge margin.
>>>>>
>>>>> That means, that in some lower duty cycle circumstances, the 
>>>>> assumed CCS
>>>>> Pd can be safely exceeded.
>>>>>
>>>>> However I cannot find an explanation that ties the duty cycle to the
>>>>> plate dissipation.  Reconciling the two data points.  Either on a
>>>>> derating or pulse basis - even as a rule-of-thumb kind of factor.
>>>>>
>>>>> The usual sources are not clear on the point.  The Eimac literature 
>>>>> does
>>>>> not come out and say it clearly.  Bill Orr loves heavy metal for
>>>>> transformers - but for SSB duty, has no problems exceeding the CCS 
>>>>> specs
>>>>> on many components including tubes.  And nothing on the net that I've
>>>>> seen links a position and some data or logic into something that is 
>>>>> more
>>>>> substantial that would pull it out of the opinion and into the
>>>>> engineering basis category.
>>>>>
>>>>> Hoping that one of you guys working in the industry - or having
>>>>> encountered this question before - may have the magic answer that 
>>>>> hooks
>>>>> the plate dis and duty cycle together...
>>>>>
>>>>> 73/jeff/ac0c
>>>>>
>>>>>           
>>>> _______________________________________________
>>>> Amps mailing list
>>>> Amps at contesting.com
>>>> http://lists.contesting.com/mailman/listinfo/amps
>>>>
>>>>
>>>>         
>>> _______________________________________________
>>> Amps mailing list
>>> Amps at contesting.com
>>> http://lists.contesting.com/mailman/listinfo/amps
>>>       
>> -------------------------------------------------------------------------------- 
>>
>>
>>
>>
>> No virus found in this incoming message.
>> Checked by AVG - www.avg.com
>> Version: 8.0.237 / Virus Database: 270.11.10/1995 - Release Date: 
>> 03/11/09 08:28:00
>>
>>
>>     
>
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
>   


More information about the Amps mailing list