-
You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
Post Pics of tube amps here
- Thread starter Lazybones1222
- Start date
First let me apologize for the pictures not coming out properly. Don't know what I did wrong but I'll probably have Big Red post them for me. Matt, A pole pig is a telephone pole utility transformer used as a plate supply. Hi Def, while a thoriated tungsten filament warms up in seconds as opposed to an oxide coated cathode that takes several minuets to warm up, it's still a bad idea to apply plate voltage simultaneously with filament voltage. It is much easier for a cold power grid tube to have an internal arc occur then one with a hot filament. On this box filament voltage is only applied after the blower back pressure switch is activated and then plate voltage comes up 15 seconds later. An over current situation will trip the grid or plate relay. Put the tube into cutoff and shut down the plate supply.
if you would like to see the rest of the pics of the wizard check out http://www.cbradiotalk.com/viewtopic.php?f=19&t=29353 he has all the pics posted in that forum same amp same numbers on front
With respect to why a cold tube can't handle high plate voltage as well as a tube with the filament on......I'm not one of those guys who pretends to have an answer for everything. In fact I'm not afraid to say I don't know the exact reason. I could only speculate so I'll spare you my opinion and speak only about facts.
I've seen one case where plate voltage was applied to a 3CX3000A7 without the filament voltage. That was the end of the tube with a loud bang and not even keyed. When I cut the anode off the tube you could clearly see where a plate to grid arc destroyed most of the grid. At the time Econco was only paying $75 for the dud and it was worth more to see the insides.
While I'll admit this is not much of a concern when the rated plate voltage is not exceeded. Many of the units on 11 meters work these tubes above 5 KV and would benefit from applying element voltages in the proper sequence. Eimac's publication "The Care and Feeding of Power Grid tubes" outlines this sequence for various tube types. It's good reading for an amp builder.
The care and feeding is good reading. It's been updated in the last decade to include MSDC IOTs. Many of the original contributors to that book are SK.
There is no reason whatsoever that this tube type will arc when cold. Internal arcing is a function of element spacing and hard vacuum. Your big blast was going to happen anyway.
Eimac does say not to draw cathode current without filament voltage. That would require keying the amplifier in any sane design. That would still not cause an arc in a good tube though.
Cathode current should not be drawn before the heater or filament has the ability to emit a space charge. The space charge prevents cathode (filament) damage. There is nothing changing from cold to warm that will make an arc happen.
Some damaged tubes will actually test OK when cold but measure grid to cathode shorts when warmup happens. I've seen this on 8877 and 8122s.
The only reason to run the filament with the plate voltage off on an external anode tube would be to getter the tube. Some of these tubes have no other way of gettering. This process would take much longer than a filament warmup timer. That's why you run GS-35Bs and old 4CX1000s for a few hours before use.
Examples of successful time tested designs that switch filament and plate on simultaneously using thoriated tungston filament tubes:
Collins 30L-1
Heath SB-200/201 SB220/221
There is no reason whatsoever that this tube type will arc when cold. Internal arcing is a function of element spacing and hard vacuum. Your big blast was going to happen anyway.
Eimac does say not to draw cathode current without filament voltage. That would require keying the amplifier in any sane design. That would still not cause an arc in a good tube though.
Cathode current should not be drawn before the heater or filament has the ability to emit a space charge. The space charge prevents cathode (filament) damage. There is nothing changing from cold to warm that will make an arc happen.
Some damaged tubes will actually test OK when cold but measure grid to cathode shorts when warmup happens. I've seen this on 8877 and 8122s.
The only reason to run the filament with the plate voltage off on an external anode tube would be to getter the tube. Some of these tubes have no other way of gettering. This process would take much longer than a filament warmup timer. That's why you run GS-35Bs and old 4CX1000s for a few hours before use.
Examples of successful time tested designs that switch filament and plate on simultaneously using thoriated tungston filament tubes:
Collins 30L-1
Heath SB-200/201 SB220/221
what about something like an AL-1200 with a ceramic tube or does the collins use a ceramic i forget
Hi Def, I never said it was a requirement to apply plate voltage after filament voltage. I simply said it was better practice to allow a few seconds of warm-up time before application. Are you suggesting it's a bad idea to confirm blower back pressure and heater voltage before applying plate voltage? Another thing that confused me was this line "Eimac does say not to draw cathode current without filament voltage". How is it possible to draw cathode current in the absence of thermonic emission on a cold filament?
That could be part of the problem since there is no electron beam to focus the flow of current from the cathode to the anode on a cold tube. Keep in mind none of the amplifier examples you have given use any more then 3.6 KV on the anode. The supplies aren't even big enough to require soft start. Chances of arcing at this voltage are greatly reduced. Name one broadcast transmitter operating at above 6 or 7 KV that applies plate voltage instantly?
The 3CX3000 failure I was talking about happened in a transmitter that was working perfect for quite some time. At some point one of the mercury contactors that initiated the soft start power up failed. The transmitter was off but the mercury contactor was shorted and holding AC on to the plate primary through the soft start voltage dropping resistor.
I heard a loud bang every couple of seconds coming from the transmitter room. When I walked in I could see the plate voltage meter up and every time it arced the voltage went to almost zero. Just repeating the process over and over until I shut down the main breaker. Opening the door on the air plenum below the tube revealed the tube was arcing inside and could be seen through the ceramic insulator.
The tube was not soft and had been making full output even at reduced filament voltage up until that day. When I build a home brew amplifier I install many parts that are not essential to the production of RF power. They are there only to increase reliability and safety. I could build a legal limit amp with less then 50 parts too. It wouldn't include a single component to protect the expensive output tube but it would work like an SB-220.
A cold filament has no thermonic emission and therefore no controlled current path. It also causes the cathode impedance to rise towards infinite. At extreme anode voltages under these conditions, the grid begins to look like the path of least resistance. The chances of a grid to anode arc are now possible depending on how high the anode voltage is.
This is not something to be concerned with in most applications because there is sufficient spacing between elements to handle the voltage even cold. The problem arises when you run extreme plate voltage. One example is certain types of pulsed modulated systems will work virtually the same tube at 3 to 4 times it's normal anode voltage. You won't want to do this on a cold tube.
One could simply say "since I'm never going to exceed normal rated plate voltage I can ignore this phenomena". For the most part I agree. Except that once you know this possibility exists, to me it's worth the few dollars on a time delayed relay to reduce the chances even further. It's called overkill.
Sorry, but I don't buy any of this.
The "current path" of a warm tube is from the plate right through the grid to the cathode. There is no difference whatsoever in potential between the plate and grid from cold to warm. Tube arcs are usually caused by gas molecules or extreme voltages, currents and sometimes localized heat found during parasitic oscillation.
Maybe we are getting somewhere though. You mentioned cathode impedance of a cold tube. This is true. Drive should never be applied without plate potential. I could certainly understand a tube going bad with drive and plate power without filament burn.
Most ceramic tubes can easily withstand pulses of 3 to 4 times their typical operating parameters. A chickenband amp running 30 or 40% extra plate volts isn't going to launch a good tube until the elements are overdissipated.
Industry standard measurement with expensive ceramic power tubes is use of a hi-pot tester to check for gas. This test is done cold. If the path from plate to grid were indeed dynamic that test method would be invalid.
The "current path" of a warm tube is from the plate right through the grid to the cathode. There is no difference whatsoever in potential between the plate and grid from cold to warm. Tube arcs are usually caused by gas molecules or extreme voltages, currents and sometimes localized heat found during parasitic oscillation.
Maybe we are getting somewhere though. You mentioned cathode impedance of a cold tube. This is true. Drive should never be applied without plate potential. I could certainly understand a tube going bad with drive and plate power without filament burn.
Most ceramic tubes can easily withstand pulses of 3 to 4 times their typical operating parameters. A chickenband amp running 30 or 40% extra plate volts isn't going to launch a good tube until the elements are overdissipated.
Industry standard measurement with expensive ceramic power tubes is use of a hi-pot tester to check for gas. This test is done cold. If the path from plate to grid were indeed dynamic that test method would be invalid.
You are entitled to disagree with me Hi Def. It seems like a very minor point to me anyhow. I like to take extra steps to avoid problems I've seen in past experience. I've worked on enough broadcast transmitters over the years to have noticed none of them apply plate voltage instantly to large ceramic tubes. Even on directly heated tubes these transmitters always delay plate and screen voltage by at least the time it takes to reach sufficient blower back pressure and a hot filament.
With respect to high pot testing on a cold tube being invalid based on what I've said, this is not true either. If a tube high pot's at 7 KV cold, that will be the worse case scenario. How about we just respectfully agree to disagree on this topic. There are plenty of other topics relating to high power amps I find much more interesting and educational. If anyone has experience with regulating screen voltage in high power tetrodes, this is an area I'd like to exchange ideas over.
With respect to high pot testing on a cold tube being invalid based on what I've said, this is not true either. If a tube high pot's at 7 KV cold, that will be the worse case scenario. How about we just respectfully agree to disagree on this topic. There are plenty of other topics relating to high power amps I find much more interesting and educational. If anyone has experience with regulating screen voltage in high power tetrodes, this is an area I'd like to exchange ideas over.
Last edited: