Print

[KB1UHP]

Hi All,
I recently got an FL-2100B off of ebay, it looks good inside and out. Yesterday I hooked it up after wiring it to run at 117v. I only have 20amp fused, 120v outlets in my location. When I started it up and adjusted it I was able to run on 20m and had a conversation with someone out in California (i'm located in Connecticut). I'm running this with my FT-450 that I have the PTT going to a relay box for the amp's RLY connection.
Here is what happened.

-The whole time the the amp was running I noticed the plate voltage was over 2.
-Amp was pretty hot, however i've never owned a radio tube amplifier so I do not know how hot this particular one is supposed to get... I do own many guitar tube amplifiers.
-After having the radio keyed for a couple of minutes the tubes got glowing red hot and one of them imploded. The output coupling capacitor, C222, was also blown, I suspect from the extreme heat that burst the tubes.

I am wondering if I had wired up this amplifier properly. All the threads I have read indicate how to properly wire up the amplifier for 220 operation. I do not exactly know, despite following the diagram for 117v in the manual, if it is the proper way I should be wiring up this amplifier for my application.

Any help is greatly appreciated!

-James
KB1UHP

[AD4U]

This amp runs 2400 volts on the tubes (2.4 on the meter).  Since the voltage was "over 2 " as you posted, that was correct.

Since you said this was your first amp, are you sure it was tuned up properly?  What was the power out on your watt meter?  It should have been around 500-600 watts. 

If the amp was not tuned up properly, it will generate A LOT of heat as you described and the tubes could implode or melt.  If this happens, other components could be taken down as well.

Dick  AD4U

[KB1UHP]

Stupid me also did not put my watt meter in line with the amp, I was anxious to try it out. I had adjusted the plate control for the dip in plate current in the manual but looking back on it I did not reduce the output from my FT-450 and gradually bring it up.  I should have taken my time 

-James
KB1UHP

[AD4U]

Hold on.....Not necessarily stupid.  From what you described, it sounds like you tuned the amp properly, but without a watt meter in line, we will never know.

The amp is designed for 100 watts in for around 600 watts out, so I don't think you over drove the amp with your rig.

The amp may have been defective from the get - go, but we will never know.

The good news is the amp can be repaired.  Obviously you will need a new set of 572 tubes and the capacitor.  From your description I am not sure which capacitor went up in smoke.  I assume it was the plate blocking capacitor or maybe a tuning or loading "padder cap".  I am not familiar with the innards of this amp.

A local HAM who is familiar with amps should be able to help you fix it.

Dick  AD4U

[KB1UHP]

I already opened her up after it cooled down from the tubes nuclear meltdown last night lol.  One of the tubes has a nice inward pointing hole from the glass melting, and the output coupling capacitor which bridges between the two tubes essentially blew away from its solder joint. Ill be ordering another pair of 572 tubes and the 'door knob' style capacitor for a replacement (located C222 on the schematic for anyone reading this in the future) I've read that that particular capacitor gets affected by the heat over time anyway and it could have just been during my use that it also failed.

James KB1UHP

[AD4U]

If it is the plate voltage blocking cap you are replacing, I would go with a doorknob type 1000uuf at 5KV.  It won't look like the Yaesu cap, but it will actually be better.

Install the tubes and power up again without any drive and without keying the amp.  HV should be around 2400 and Ip (plate current) on the meter should be zero.  Obviously there should be no abnormal sounds or heat or smoke.

If everything is OK so far, install your wattmeter after the amp and connect the wattmeter  into a dummy load.

Key the amp with no drive.  The idling plate current should be around 90mA. 

If OK so far I would set the amp for 80 meters to start with.  Preset the tune and load capacitors as indicated in the manual for 80 meters.

Apply a LITTLE (10 watts or so) drive and see if the plate current goes up.  Watch the wattmeter to see if power is coming out of the amp.

Adjust the tune and load controls for maximum output power as indicated in your wattmeter.  With 10 watts drive you should get around 50 watts out, if everything is OK.

Carefully apply more drive from your rig and adjust the tune and load controls again for maximum power as indicated on the wattmeter.

When finished and with 100 watts of drive you should get around 500 - 600 watts out of the amp with around 500mA of plate current as indicated on the meter.

Do the same for each band.

If all checks out, you are ready to go.

I recommend that you always keep the wattmeter in line.

Dick  AD4U

[KB1UHP]

Thanks a million for posting that procedure, Dick.  Greatly appreciated. I was eyeing a 5KV capacitor as well to give me a little more 'headroom'.  Ill let you know how it goes when I get the parts in.

-James
KB1UHP

[W8JI]

Thanks a million for posting that procedure, Dick.  Greatly appreciated. I was eyeing a 5KV capacitor as well to give me a little more 'headroom'.  Ill let you know how it goes when I get the parts in.

-James
KB1UHP

James,

If you melted a hole in the glass, it is 100% your fault. :-)

You should always have a power meter in line on the output.  Even 20-30 watts of drive can blow the amp up if the amp is not tuned properly.

You might want to read these links:

http://www.w8ji.com/loading_amplifier.htm

http://www.w8ji.com/fl2100_problems.htm

73 Tom

[K7ZRZ]

James,

A couple of other things you should inspect very carefully are the plate choke (big multi-section coil between the tubes) and the suppressor resistor/coil assemblies (right at the plate HV caps). It seems possible you could be experiencing an oscillation at some other frequency, which could cause the tubes to be dissipating a lot more power than you would normally see if that wasn't happening.  You would be able to determine if that is the case by watching the plate current carefully, which would rest considerably above your normal idle current of 90 mils when you stop talking.  I think an oscillation would be more likely to happen on the higher bands.

There is something about a modification to the bias circuit, that you should carefully investigate, when replacing tubes with new products. The owners operating manual PDF file I have contains a section about that. And my Buddy Jeff, who has mine (I've actually never seen it) performed that mod when we got new tubes for it.

Fortunately, the amp isn't too complicated, and probably a pretty good one to "cut your teeth on."  But you have to be careful. Discharge caps (or wait an hour), don't be "in a rush" - and by all means, follow the very good set of instructions that were given by Dick AD4U.  Having the assistance of another local ham who has experience would be great, but I realize that this is not always practical or possible - as in my case. Then... you need to rely on common sense, careful attention to all the things written in the manual and (some of) the great advice given here and on other forums. With patience, you will eventually get the amp to a stable condition and be quite happy with it.

Good luck,   Brian K7ZRZ

[K7ZRZ]

James,

Another thing I thought about before when I was typing the first part, but didn't get around to thinking about it twice, is that of the main power fuse on the back of the amp.  If you were drawing enough plate current to cause a tube to melt, and the main fuse didn't blow, you should find out why.  You may have a fuse in there that's a larger value than it should be (or has aluminum foil wrapped around it     ??).

Brian K7ZRZ

[KB1UHP]

Good thought Brian. The fuse is a ceramic type is a bussman abc-20. (FUSE, 20A, 250VAC, FAST ACTING, CERAMIC BODY)
The plate choke actually appears good, but I will examine further. I was surprised that it occurred after opening the amp and seeing how relatively clean it was.  I plan on opening it up and doing a more complete disassembly when I finish moving. Where I live now I just have room for my computer, no work bench or anything. The new apartment will have a room that I can dedicate for my radio, computer and guitar gear. I can't wait.
I will be back as soon as I am able to open it up and get the new parts in there. Ill keep everyone up to date.

Dick and Brian... Thanks so much for the valuable advice, its already expanded my knowledge base a lot.

73s
James
KB1UHP

James,

Another thing I thought about before when I was typing the first part, but didn't get around to thinking about it twice, is that of the main power fuse on the back of the amp.  If you were drawing enough plate current to cause a tube to melt, and the main fuse didn't blow, you should find out why.  You may have a fuse in there that's a larger value than it should be (or has aluminum foil wrapped around it     ??).

Brian K7ZRZ

[KC8ADU]

i have seen several of those big green caps with cracks in them.
replace them all.
they take a beating and do not get better with time.
i own 2 fl2100b's and both had cracks in the caps!
and fully agree that if you suck in a 572b you were doing something very wrong.
hopefully they were cheap chinese ones and not cetrons .it would suck to waste good but virtually extinct tubes.

[W8JI]

James,

You might consider my comment it was 100% your fault harsh, but it is true. I'm not one to dance around the issue. :-)

It take terrible heat to actually melt glass. This is NOT directly related to fuse size or drive power, it is related to tuning and operation. Both are under the direct control of the operator.

The dissipation or heat produced is a direct function of the input power to the tubes, and the output power to the load. If you had 2300 volts at 500 mA plate current, that would be 2300*.500 = 1150 watts of power. If the real power going into the feedline was 740 watts, that would be 1150-740 = 410 watts of heat.

Each tube in your amplifier is rated at 160 watts of steady anode heat, or 320 watts total. On CW or voice, because the load is varying and not steady, 410 watts is OK on peaks. The thermal averaging in the mass of the tube will allow heat peaks well over 320 watts. This is how the amp should have run.

On SSB voice, you must have operated for an extended time with much more than 500 watts of average heat. That is what it would take to suck the glass in on a 572B tube.  This means the peak heat must have been terrible!

This might make your head swim, but the concept is the important thing.

Someone else suggested bad components or parasitics, but in truth that is very unlikely to be a root cause. Even if it were true, watching the power output meter and the plate current would have warned you right away.

You need a good output meter and need to know how to use it, and you also need to understand how to read the plate current meter and what it means. Otherwise you will be right back into expensive trouble. 

I'll add a plate current vs. power output table on my web site to help you and others. I'll work on that today. Your problem, and the responses here, have caused me to realize something lacking in amplifier instructions. Since I design amplifiers and write manuals, I can do a better job because of reading about your problem and the suggestions of others. You spent the money, but I learned something that is missing from manuals. Every manual should have a plate current vs. power output chart or graph!!

I've never thought of this before but I can see how a plate current vs power output graph or chart would be one of the most valuable assets of all, other than telling people to watch grid current.

73 Tom

[KB1UHP]

Don't consider the comment harsh at all Tom. I rather just know what I did wrong and do things right. How else would I have known if im just sitting alone here with my amplifier burning stuff out again and again? After reading your explanation I completely agree with you.  Lessons learned and don't worry, because of this ill make sure to become an expert in the operation  with these amplifiers.

73s
-James

James,

You might consider my comment it was 100% your fault harsh, but it is true. I'm not one to dance around the issue. :-)

It take terrible heat to actually melt glass. This is NOT directly related to fuse size or drive power, it is related to tuning and operation. Both are under the direct control of the operator.

The dissipation or heat produced is a direct function of the input power to the tubes, and the output power to the load. If you had 2300 volts at 500 mA plate current, that would be 2300*.500 = 1150 watts of power. If the real power going into the feedline was 740 watts, that would be 1150-740 = 410 watts of heat.

Each tube in your amplifier is rated at 160 watts of steady anode heat, or 320 watts total. On CW or voice, because the load is varying and not steady, 410 watts is OK on peaks. The thermal averaging in the mass of the tube will allow heat peaks well over 320 watts. This is how the amp should have run.

On SSB voice, you must have operated for an extended time with much more than 500 watts of average heat. That is what it would take to suck the glass in on a 572B tube.  This means the peak heat must have been terrible!

This might make your head swim, but the concept is the important thing.

Someone else suggested bad components or parasitics, but in truth that is very unlikely to be a root cause. Even if it were true, watching the power output meter and the plate current would have warned you right away.

You need a good output meter and need to know how to use it, and you also need to understand how to read the plate current meter and what it means. Otherwise you will be right back into expensive trouble. 

I'll add a plate current vs. power output table on my web site to help you and others. I'll work on that today. Your problem, and the responses here, have caused me to realize something lacking in amplifier instructions. Since I design amplifiers and write manuals, I can do a better job because of reading about your problem and the suggestions of others. You spent the money, but I learned something that is missing from manuals. Every manual should have a plate current vs. power output chart or graph!!

I've never thought of this before but I can see how a plate current vs power output graph or chart would be one of the most valuable assets of all, other than telling people to watch grid current.

73 Tom

[W8JI]

Don't consider the comment harsh at all Tom. I rather just know what I did wrong and do things right. How else would I have known if im just sitting alone here with my amplifier burning stuff out again and again? After reading your explanation I completely agree with you.  Lessons learned and don't worry, because of this ill make sure to become an expert in the operation  with these amplifiers.

73s
-James  

I added some stuff here James, but I'm trying to think of how to make the data useful and simple:

http://www.w8ji.com/fl2100_problems.htm

I could use some ideas on how to make it useful.