50 Ohm "Bird Termaline" 82C Coaxial Resistive Load, and the Repair Needed

[semaphore1999]

Electronics folks, and especially RF gurus:

I am not an RF guru, but the industry has forced me to learn some things this direction to remain relevant. So I am hoping to poll the set of electronics knowledge at DMS and beyond to find answers to my questions about an RF item I stumbled upon ownership of, thanks to DMS freebie shelf.

Many would say its unrepairable. I disagree. And it is for a specific reason that I disagree, as outlined in the video that can be seen below.

My question are:

*Can anyone think of a way to bond the ceramic portion beneath the conductive cylinder permanently so as to fix the load once and for all? I am hopeful for mechancial robustness, but I might settle for fragile but functional.

*Are there good methods the experts know of to make sure that the electrically conductive portions of the system make low resistance electrical contact over the entire circumference of the break?

*What is silicone water as described by Bird? They use this for its dielectric properties as the resistance is submerged in this stuff I think. Can anyone tell me for sure if my understanding is correct? Where do I get silicon water? Or can I make some out of common items? How about not so common items?

*What is the fluid that gets pumped through the cooling coil? What pressure is required to get it moving fast enough to support 2500 W? How much heat must I remove from the cooling fluid outside the Bird so I can support 2500 W? Any suggestions for a piece of equipment that can do this?

*Do we have a piece of equipment that can inject 2500 W of power into the load? If so, can someone train me on such so that I can characterize and ensure that the unit I have can be used to sink this much power for an extended period of time? If no, where can I get access to such a signal source?

*From my reading about this subject, I have ascertained that with the power levels involved, an engineer must use extreme caution to avoid shooting large amounts of RF into the air around where I am using. There can be health implications involved. How do I ensure I am being safe here? What are the mistakes that others have made that are to be avoided? Can you point me to where I can get a good education on safety when dealing with large power RF signaling?

Here is a video I made that shows the internals of the Bird 82C and the damage that the unit I have has sustained:

Special thanks to our young maker friends Halee and Hannah. The intro for 3E8Tech has been created by Haylee and I think that Hannah has been involved and/or will be in the creation of high res improvments. The intro helps my professionalilsm a lot, and yes I heard you young prodigys about the snippet at the beginning, but heard it after I rendered this one.

[semaphore1999]

I will likely bump this post every so often as it is a specific sort of expert that may have feedback on the subject.

Or nobody may have feedback on the subject and I may be in the position of simply pushing the two parts together and attempting to apply a ring of solder where the break is.

Still, I don’t know what “silicone water” is. What in the world?

[wandrson]

Still, I don’t know what “silicone water” is.

They are probably referring to Sodium Silicate solution aka water glass

[Raymond]

I will likely bump this post every so often as it is a specific sort of expert that may have feedback on the subject.

How about posting some nice, clear high resolution pictures?

[tmc4242]

I think you want Silicone oil. Not water.

That unit is rated 500W continuous. Not 2500. Duty cycle at 2500W is very small.

We do not have a 2500 W RF source.

Step one though is a DVM. Should measure 50 ohms. If not, do not apply RF.

Book says replace RF section as a unit.

No time to watch videos.

Not sure what you’re trying to fix, but if it’s leaking there’s a specific part number O-ring that might help. See the manual.

[semaphore1999]

@tmc4242, this is not the good, working 500 W unit in the electronics room. This is the broken 2500 W liquid cooled unit that is my personal possession (dismantled in my car right now) as I obtained it from the freebie shelf months ago. I think that there may be two fluids involved, dielectric fluid and the fluid that fills the cooling flow lines on Bird Termaline 82C (the 500 W unit does not have these cooling lines).

But of course, the video explains these things. Its easy to form an opinion based on the general look of things, make some misguided assumption about what the message is that I am presenting here, then respond all wrong out in the open for all eyes to see. I would know, I do it all the time. However, the questions here will have to be addressed by a bona-fide RF expert AFTER examining the questions carefully, which, I am afraid, means you gotta watch the video. More and more of my posts will go the video route since a picture tells a thousand words. I use 30 fps, so this particular video replaces a post of mine containing 1,107,000 words. You can find several posts of mine of this length 4 sure prior to my incorporating camera use, so I regard video communication as improvement… and I might even work on such feedback I have received that tells me I need to do less talking and a lot more action on the screen (sounds like spot on feedback, but gonna be really, really hard for me… years of bad habit reinforcement blabbering on in front of audiences as the various technical experts I used to play on TV…)

@wandrson @Raymond, I will read about the water glass and//or any related chemical cocktails to see which one has the notable dielectric properties. I think it should be possible to isolate it this way. Bird selected this fluid due to its ‘dielectric properties’, according to the Bird 82A manual (I found a 82 C manual in another language months ago, and translated. Sadly, the copies were not up on onedrive, so I lost them when I crashed my harddrive). I think that the 82A is like the one in DMS lab…its 500W and has no liquid cooling ports. Which begs another head scratcher: what fluid do I use to cool the device? The cooling fluid is kept separate from dielectric fluid, and it runs through a little coil shaped pipe that circles about the coaxial resistance, over and over until it reaches the end of said R, then comes back to the front of device and back out an output port shaped similarly to input port. It also keeps a fairly uniform distance from R.

I will pull things out and get some better shots of the internals and explain my proposed fix, Its just a solder job, the more I think abiout it, and should go well as long as I can get a nice, uniform ring of solder along the circumference of the break (which makes no sense until I get you those photos).

One last thought rolling around in whats left of my brain after months of auto work restoring one Mazda Millenia, and all the falling on my ass making a new and different mode of electric transportation, which turned out to be more of a recreational/educational product? thing perhaps…no I really don’t know for sure what/how to do with it, and want some help there too cuz pullski is hella fun to ride around on and everybody who has tried it LOVED it. Also, they did not fall, as nobody but me has done that, and this only occurred PRIOR to brake installation, Now that there is a brake on there, not a single rider of any age or usage method has fallen! And many have rolled around pullski powered. Also, folks can just grab it and go. With no training, no experience, they just take off and go like an expert would. Its a very natural feeling, easy to use, safe, and possibly an excellent training tool for learning to skateboard, rollerblade and who knows what else one can do with it. Alas I digress, this post is about Bird 82C, of which better pics are coming soon to assist in understanding the device and my questions. Thx all!

…oops, this post is worth at least a pic or two…lol…

[tmc4242]

I did not reference the one in the lab.

You specified a Bird Termaline 82C.

The manual for same says 500W continuous, higher at reduced duty cycle.

Sorry that someone who is as prolific a producer of words as yourself cannot succinctly produce a written description of the problem.

Best of luck.

[TBJK]

You sure it’s not transformer oil?

[semaphore1999]

@tmc4242

The 82C is 2500W, please confirm with Bird as my post is not a testing ground for flat earth or conspiracy theories.

I do see why you think that because some yahoo made an ebay ad and mixed them up. But I have been fiddling with thing for a minute, and can assure you 82C is 2500W and 82A is 500W. Simple. It also helps that I have access to both types of unit, and they are marked right on the outside concerning maximum power.

Problem Statement:
To fix broken Bird Termaline 82C 2500 W 50 Ohm Resistive load.

However, as the engineer posing the question, and also working toward higher quality videos and editing proficiency, I think I will be the one who selects my communication format. I find that most prefer a video as it is a faster, and I think better mode of communication. It leaves less to the imagination which is double edged sword. Good in the sense that I can be much clearer about what I intend to communicate. Bad in that I do not employ the imagination of my viewers/readers. For engineering/scientific work, I think both the good and bad are plusses for video. I hope to see, and think that I will, many others follow suit with their work,

I have a whole bunch of cameras that I picked up off a favorite getting spot of mine. @tmc4242, I would give one to you if you would like and it helps you maybe experiment with newer method of communication. I can even recommend editing software, although I am still a pup in this arena, hopeful my young duplicate friends can grow me a bit…lol

pics coming for those who want a clearer view…if I can get done validating that the earth is round…

[tmc4242]

Actually, Google led me astray on the first pass. 2500W point conceded.

I prefer reading over watching videos because I read much faster than most videos go. Communication in that format is much clearer too, assuming the author has sufficient command of the language in question.

Make what you like, show it how you like. Not my problem.

[semaphore1999]

You know, its interesting because I grew up attending a private school in East Texas, and for those who attended All Saints, we knew our parents all paid more for better instruction. So in the afternoon, religiously after lunch, when the eyelids are heavy and our developing systems were carb loaded with macaroni and grilled cheese, pork and beans, whatever high starch/high fat/high sodium/ high cholesterol diet settling in our bellies, the powers that be had sectioned off the first few minutes after lunch as a proper dedication time to a good, informative, educational color film that they had located at the library, using the Dewey Decimal system, no doubt. I time, a few minutes hadbecome one after lunch, and with the encouragement of lazy kids with full bellies was for sure that there should be no limits placed on the length or number of films that could, excuse me SHOULD, get played during that hour, extending beyond that hour as need be for the benefit of our education. And so the films rolled, and us lazies reveled in our naughtiness, and the brilliance we had clearly employed avoiding the hard work those dreaded teachers wanted to do with that time instead.

I tell you this however because it’s important to recognize what happened to all of the students going to All Saints as we watch those films early in the development of our cognitive abilities. The film watching went on for literally years, perhaps my first grade student Year to my six. And again this is early and development cycle I had just learn to read and just learn basic arithmetic and now was watching films everyday science-based, many of them demonstrating three-dimensional rendering.

For this reason I developed an appreciation for film and feel that it is a faster medium for learning. There’s definitely more information available to the consumer of the data as one can argue about the megabytes of data flashing in front of the screen. Now we realized and a lot of it is just duplicate colors reinforcing the pattern being displayed. But nevertheless there’s a lot of data there whereas with reading it is a a more serial stream, and considerably slower. I think that there are advantages to either style of data transfer, but the advantage for the slower stream is really surprising and interesting tribute to the flexibility of the human brain. And that is the fact that as we get our serial stream in from reading or listening to an audiobook our brain is more idle and we have time to imagine things the way that we might want to imagine them. It’s sort of a creativity that gets mixed with that serial stream and humans make it to differing interpretations of a given story or given data exchange the because of this.

So I will accept the fact that you are in no hurry to change to video, however I am hoping that you will accept my decision that I want to push my broadcast capabilities in that direction, as I don’t want a lot of imagination applied to what I’m trying to communicate. I’m an engineer and I want to completely specify exactly what it is I’m communicating and every facet that there is about it, leaving no room for imagination to skew the presentation I would like to present.

With this, yes i realize iam no pro at videocrafting yet, but working on it, which means i am learning. So i did also hear the request from Folks at DMs that they wanted to see the inner workings of this dummy load high power resistance oh, and it was very interesting so I think all share the photos that I took last night wall covering he door at the Malibu Bar. Notice that I can measure 50 on across the cylinder of material grown around the ceramic. The copper hanging out the ends of that 50 ohm cylinder, have near zero resistance and or a great place to probe. Furthermore they are a wonderful place for me to attach the input conducting shape that is basically the middle part of the coaxial connector coming into the device. In other words all we need here is a conductive connection to the 50 ohm resistance and then to re-suspend the 50 ohm resistance in the dielectric goop. From there I have a strong belief that we can dump 2500 ohms in there as long as we set it up straight and run the fluid through to keep it cool. Certainly I would want to do this testing, but that is a discussion for a different day. For now I would like for anybody who has input on this subject as to problems that they see with what I am proposing here, or maybe suggestions or improvements I could make that will make it work out better. Meanwhile check out the pictures below to see better what’s going on with this dummy load:

(And since this post got really long with all the pics, I’ll follow up with my concise list of questions ijn a minute.

The fact that the input center conductor piece has rolled into the side of the 50 ohm piece is nothing more than confusing cooincidence unrelated to scope of the effort here. While the tip of that conductive input piece could indeed affect a resistance measurement, the fact that it is going nowhere fast utilizing the ‘T coinfiguration’ saves the day and my measurment, which was very tough to do while holding a camera, lol

[Raymond]

Perhaps something in a high heat conductive epoxy?

[semaphore1999]

If ulook down into tbe near solid cone shaped input peace, you can see that it is hollowed out for maybe a centimeter from the breakpoint. It looks like it may have been broken and repaired once prior to this break. It looked like they attempted some sort of epoxy, but the epoxy has broken down and whatever silicon water fluid is used. So the epoxy has become this black mold like goop and the ceramic still retains whatever form it is after the break.

My proposed solution is the following: I don’t want to make any changes to the ceramic. I don’t feel that the deviation in ceramic which I believe will be cracks of less than a couple of millimeters worst case, but usually almost non-existent once the thing is held back in its original shape. So I don’t think there’s significant dielectric disturbance even with the crack still there without epoxy. So my plan is very simple. I am just going to push the thing back into its original state, of course after cleaning the black goop off of the ceramic. Then I am going to use a good soldering instrument and solder all the way around the circumference of the 50 ohm impedance and its input cone shaped entity. I want to be sure that I form a solder Bridge around the complete circumference of the break. At no point should there be a break in my solder Bridge. In other words there should be a 360-degree solder Bridge formed from input cone to 50 Ohm resistor input ring. I believe that this solder solution will make a solid mechanical connection between imput cone and 50 ohm resistance.

The only concern that I have remaining now that I have established in my mind that this should work is the following. How hot is it going to get inside there and will my solder melt due to these temperatures?

[kbraby]

2.5KW is a very non trivial amount of heat. It is a strong instant hot water heater. If I was trying that power level, I would be running the cooling line with cold tap water and dumping the hot water down the drain. It is quite possible due to the thermal conductivity of whatever dielectric fluid you have in there that it will be much hotter than the exit temperature of your coolant flow.

Remember that if the characteristics of that joint change due to melting, or other factors, that joint is at risk of reflecting some of that power back to the source, instead of passing it to the load. And most RF amps don’t do well with a meaningful portion of their output reflected back.

Is it possible that the only difference between an 82A and an 82C is the cooling coil? Would it maybe be safer to source an old 82A as a donor for the dielectric and complete load element/input cone?

[semaphore1999]

Clever idea. Not sure who woukd wanna give up their functioning 82a tho., lol… Unless it upgrades them from 500 to 2500. Hmmm, who could that be?

Solder gets pretty hot b4 it melts tho, and the whole thing is submerged in fluid. I would also think that thermal convection occurs in the fluid, draws heat away, but dont know 4 sure. Maybe a test case with ac straight from wall gives us a feel for what 115 rms does, then multiply the effect by ratio of powers from 250w (115vrms approx) to 2500 w? So 10x increase. I could drop a thermister in there, probably wouldnt change things too mich if xleverly placed.