The AM Forum

Hello All,
Is anyone familiar with the Ten Tec Titan 425 linear amplifier??
I have an arcing problem in the tank circuit. As I tune and apporach full rated power the arcing starts.
I'm trying to avoid bypassing the safety interlock and operating with the cover removed to see the source of arcing. But might have to be done to correct the problem.

I have been doing searches on the net for any track record of this problem. Seems that the main concern is the final RF pubes. Very expensive to replace and easily abused from overdriving the grids.
Thanks

Fred

You'll probably need to do that MOP. I don't know how else you can see the source of the arcing. Have you opened it up and looked for signs of arcing (carbon, zorch marks, etc)?

YUP Steve
Looks like the path of adventure to take. One hand in the pocket and all that.
It did this before and I opened her up and did not see any signs of the arcing. I'll have to fire it up and look at live conditions this time.

Fred

Chances are pretty good it is in the load air variable.  It is looking into a bad impedance that has changed on you recently.  Same thing happened to me with my AL 1200 into my 160 meter inverted L.  Turned out the radials had enough corrosion at the attach point to cause impedance problems, hence the arc at near full power.

Bet it doesn't do it into a good dummy load does it?

Just my 2 cents worth Fred, Don't bypass the safeties. I really think if you take the time with a really bright work light and a magnifier visor you can find the problem. If you can't do that I recommend you put the amp on the bench and find a way to look in through the fan vents. You might find the direction of the arc. Than open her up and investigate.

I know you know the interlocks are for your safety. I honestly think some patients and time spent searching will pay off.

If you use the interlock bypass method Please build a bang stick first to discharge the caps and be really careful please.

Mike

Yea Mike,
I think you can touch 2500vdc one time.

And the arcing is with the amp in a known good dummy load.

I'll have to remember to bring my magnifier visor from work and smoke it over this weekend. It's been convenient to have the Raytheon on 1885 and the linear set-up for 1985.
Thanks

Fred

Don't know about this particular amp but does it have switched in lumped capacitance in the output side of the tank?  One or more of them may be bad. 

It's not like 2500 volts is going to jump out of the box and get you. You can safely bypass interlocks and remove the appropriate covers to see what you need to see. Stand back and watch for the fireworks. Dimming the lights in the room can help to see the initial area. Then you will know where to look. I'm not recommending sticking your hands in the box, just standing back and observing.

Ok
I'm making a list and I'm gonna check it twice!!!
Safety, observing the source of arcing and looking into the bandswitch. There are definitely caps switched in for 160M. And they were changed out as a mod from Ten Tec for "a change in tuning while transmitting problem". Not an easy amp. to get into. Tight spots.
I'll get back to you later this weekend

Fred

Actually, it would be possible for me to touch it a second time but somehow I don't think I would get off so easy if I did.

Been there.....

http://webpages.atlanticbb.net/~w3np/hvsafety.htm

Trust me Fred, it ain't a fun experience!! I would still open up the Titan and do as Steve suggested. Watch it in a dimly lit room for anything amiss. I was just inside my 812H/V70D rig today. Like the carpenter that measures twice and cuts once - I carefully chart my moves and the next position of my hand.

Take care.

I had one of the switched caps arcing in the 40 meter bandswitch position.  From there the switch added more fixed caps to the variable so tuning was off on all bands from 40 meters down to 160.  That is how I isolated which cap to replace.

I found it by doing like Steve suggested, then used a 10 watt 100 ohm resistor and made a grounding system out of it.  The HV in my amp is 3600 volts but anything above 1 KV is VERY dangerous. 

I learned to work on HV stuff under a master who taught me on a 1 KW system in a BC transmitter back in the 60s.  It probably will be something dirty or shorted like the cap in mine.  Do exercise caution.

There are definitely caps switched in for 160M. And they were changed out as a mod from Ten Tec for "a change in tuning while transmitting problem"

phred, there's your problem.  8)

I know the 160 meter fixed padders were a problem for the original amps, I looked into buying one before I decided to go with a homebrew amp instead.  The originals were not capable of handling the current and the loading would shift enough that during long dashes it was evident.  I presume they upgraded those but you might have an earlier model in which the caps have died from "heat prostration".

You should be safe enough observing although it would be a good idea to wear safety glasses in case it ends up sending debris out from the arc.

Rodger WQ9E

Here's a little report on the Titan.
Looking down inside the first section of the bandswitch is arcing within itself.

According to the sheezo, there's a lead coming from the RF final pubes to this section of the bandswitch and there are some transmitting caps that are switched into the tank. I'm thinking that these are the loading caps?

This amp had the mod loading caps installed before I bought it for that drifting problem.
I have noticed that the TUNE (PA Tune) cap is fully unmeshed for tuning at 1985. The loading cap is half meshed. 2000 watts out 70 watts drive.

THanks for the advice and guidance
Fred

fully unmeashed is a problem. You are on the wrong band man

Actually the Titan has a 160A and 160B selection on the bandswitch.

I'll try for grins what 160A does at 1985. It worked about a month ago on 1985 using 160B. Full P.E.P. no arcing. All tests and set-up are done in a dummy first before going live into the antenna system.
I'm not getting passed the Dummy load.


I'm assuming that "B" is for the upper part of TopBand. The tuning chart seems to indicate that TopBand was broken into segments, prolly for the extra loading required on the lower end 1810-1850????

Fred

It doesn't look too good in the 'ole Titan. The tank circuit is a Pi-L network. The transmitting caps for the Pi side all have cold solder joints. I scanned and posted the scheezo. The bandswitch that is arcing is S1C.
This has a group of transmitting caps with cold solder joints. I can't even begin to think of getting a solder gun or solder pencil in there.
The S1B has a transmitting cap that the phillips screw is completely broken away from the capacitor.
The transmitting caps look like good ones. There is also a pic of some of them on here.

Using the ceramic tubes encourages the sleek design and a very hard device to get into to repair. I'm glad the Raytheon is here. There will be hassles for large QSY's, but I guess that's life.

I'm sure Ten Tec will charge a healthy price for repair/rebuild of the tank circuit. I know I don't want to get into it. Unless I get real desparate for operating.

Thanks for looking. Has anybody worked on TenTec stuff??

Fred

There is only 100 pf difference between 160a and 160b for loading. the 400 pf extra on the plate tune is in for both A and B. No way should you be needing to run the loading fully unmeashed usless there is a vswr problem.

I bet the tuning knob setting is due to the other problems with the fixed loading caps; until you get those straightened out nothing is going to behave normally with the output tank.  It is also quite possible the 400 pf unit was damaged when problems occurred on the output side of the tank. 

Ten Tec is supposed to be pretty good about service so I would get in contact and get a quote.  If there are cold solder joints it sounds like that happened during a "field upgrade" of the fixed capacitors.

The only work I have done on Ten Tec is repairing the PTO's in my Century 21.  Hopefully if Ten Tec pricing is not reasonable you can find someone close by to help you out.

Rodger WQ9E

Phred,

I'm surprised by how ten tec has those padder wires arranged - that is at least 3 zorches waiting to happen. ( or that has happened)

if you trust a one handed brain damaged cancer patient to repair it, ship it here and I will look at it for nothing. It looks rather straightforward. if I don't think I can fix it, I'll do nothing to it and  you can arrange to have it shipped to Ten Tec from here.

note: I know thats a high dollar item, NP if you would rather have someone else do it.

Ususally the design impedance for the juncture of the L coil/load cap to the Pi coil is 300 ohms.  If the amplifier is seeing something other than 50 ohms at the antenna relay, then this is reflecting back to S1C which is where the problem is.  

Are you sure of the contacts in the antenna relay?  What about the wire connecting the L coil to the antenna relay?  The cold solder joints could be the result of high circulating current which heated the solder.  From the picture it appears one of the caps arced which may have been the cause of the problem, then others.

I would bet you had some sort of antenna trouble and when it arced the first time, it did the deed to the rest of the circuit.  I suggest you do your own repairs.  The expensive thing will be the doorknobs.  Those things are in the 25 dollar range, and that is each.  Just think of it as a challenge and take your time being sure of all mechanical and solder joints.  High power presents its own problems at connections.

When I had to replace the two doorknobs in mine, I bought from Ameritron.  Turned out Surplus Sales of Nebraska had them cheaper than Ameritron.  You might check there if you do your own repairs.

Maybe an off the wall idea here, but did you try to contact Ten-Tec Phred? They have a Tech Support group that is supposed to be second to none. What the hell give them a try and ask them where to look. I've heard they have sent parts without an order.

I hate to throw the white flag up on this, but it requires a lot of disassembly before even getting to the problem area. Placement of wires and the proper doorknobs might get frustrating. I would get pretty hot if I went through all of that mechanical rig-ama-role and the dam thing  is worse off, than when I began.

I'll check with TenTec. Like Mike(y) says and, maybe one other, TenTec is suposed to have quite a team in the repair shop and they will give it the final smoke test before leaving the shop. It's a pretty strong amp and not afraid to put out the P.E.P. Dead carrier was a little over 2000 watts and I was not driving the tubes very hard, about 70 ma. I would hate to pop expensive ceramic tubes like those with my hit and miss approach. They are supposed to last a very long time.

I'll give an update what TenTec says they'll do for me. Thanks for the great input, though. If it was more of a traditional design and not using ceramic pubes, I would jump in there with a Weller D650........300watt soldering iron and do my thang

Phred

Phred,

Knowing when not to tackle something is a very wise; too bad it is a lesson lost on many politicians and business leaders.  You definitely don't want to blow up anything expensive (or irreplaceable such as the operator!)  I have toured Ten Tec a couple of times while down in the Smokies and it appears to be a very friendly group.

A good trick for checking the output network after working on an amp and before putting in the expensive tubes is to use one of the generic antenna analyzers like the MFJ 259 series to check your work.  Since the output network is an impedance transformer it works equally well in both directions.  So calculate the normal operating load for the amplifier tubes and then connect from the anode connector a resistor (to ground) of the value calculated.  Of course all of the testing is done with the power disconnected.  Connect the analyzer to the amplifier output and either bypass, manually close, or supply power to the T/R relay only (whichever method is easiest).  If everything is as it should be the analyzer will indicate a match with the bandswitch, tune, and load controls in their proper position.  This won't catch parts that will arc over at normal operating power but it will catch wiring, soldering, grounding, or component value errors before you turn dollars into smoke. 

The 425 is a nice amp and I hope yours is back to making RF soon!

Rodger WQ9E

Try loading iy up on 75 and 40 and see if it plays. The 400pf cap open you might be fully unmeashed to get a dip on the second harmonic. The caps still look ok  so maybe the band switch is bad.
I agree with Derb those 3 cap wires look pretty close. 

the middle door knob has a huge zorch and I bet it runs all the way through the ceramic. Whomever did that work has not a clue of he nature of HV.

I'm gonna try one thing before considering TenTec......I'll try to get a hot pencil type iron on the three doorknobs on S1B and get to the cold solder joints, THEN I'll get to the one dooknob, that I marked with red, on S1C, and solder the phillips screw term back on to the top. Can this cap get hot to melt solder?

I am familiar with the MFJ 'looking back' back into the tank. You have to disconnect the pubes and connect an equivalant value resistor to immitate the plate resistance.

Thanks for the tips. It will make RF one day.....well it does now, but with a little electrical show.

Fred

You don't have to disconnect the tubes Fred.  Just hook the resistor across them to ground. 

If those doorknobs have been zorched or near one that has, then replacement of the zorched one is necessary.  A very thorough cleaning of the others is in order, including in all crevices.  Some of that zorch matter can be like the carbon tracks on ceramic switches.

When you finish, the thing should shine a much like a new baby's bottom as possible.

Jim is correct, the tubes need to be in so that their output capacitance remains in the circuit.  It doesn't make a whole lot of difference on the low bands where significant plate tuning capacitance is used but it will throw the results way off on the upper frequencies if you have tubes with a fair amount of output capacitance.

Given the price of 3CX800 tubes I would seriously consider getting those other door knobs out and getting them to someone who can "hipot" them.  When the zorch event happened you could very well have collateral damage and until you test you won't know whether hidden damage in one is likely to also cause damage to your new components.

I would also take a look at the RF choke on the output side (used to create a short to ground for DC appearing on the output tank) to make sure it wasn't damaged.  I have seen these chokes develop internal shorts across a number of their turns after other amp pyrotechnics and this will create some very odd tuning behavior.  They are pretty cheap so if there is any question replace it since it is also a safety device.

Rodger WQ9E

Believe it or not I was able to get to the cold solder joints and soldered the philips terminal back onto the the doorknob cap. From my reading, the plate impedance for a 3CX800A7 is 2.7K. There are two in parrallel sooooo, take one half of that 2.7K????
I would like to connect the MFJ259 looking into the Plate Tank to investigate the fully unmeshed TUNE variable cap in the 160B for the upper end of Top Band. 1880 to 2.0Mhz

Thanks..I"m trying to take a stab at this before sending it to TenTec.

fred

Plate load impedance is roughly 2 times the plate voltage Fred.  Or if you want, EP/2 X IP or as the old book say, EP/1.8 X IP

Be very sure everything is CLEAN and TIGHT before you fire it up.

Fred,

Are the tubes operated in Class B in the 425? If so then the load is voltage/1.57 times plate current.  For AB1 approximate value is voltage/1.4 times plate current.  This load is based upon the total current so just calculate this way rather than considering the number of tubes.

Rodger WQ9E

You are right Rodger, I am thinking AM (class C).  This isn't. 

Jim,

This is more of a class C board  ;)

Rodger WQ9E

The manual says class AB2.
I've had enough play-time. Contacting TenTec tomorrow. Everything looks better with some soldering but S1C still arcing under full power.
I did the MFJ259 looking into the Tank and it is happy. The TUNE cap is now about 3/4 mesh for tuning 1985.
Thanks for the great tips and reminders how to use tools already on hand.

Fred

Don't be so shy about calling TenTec.   Every time I've called them they've given me right on the money free advice.   They have a lot of experience with the 425.   When I had a 425E and blew it up by routing the output of my class E into the the input of the Titan and then keying everything up, there was a fellow (I can't remember his name) who on series of phone calls led me through diagnosing the damage.

I got it working again.  They sure made a fan out of me.