fixture for opening RF solder can?

[coppercone2]

So I got a little used to resistance soldering and brazing and using a big resistance transformer to heat things. I want to open a soldered together RF amplifier, but the problem is the geometry of the part. I thought to glue a pull handle to the top of the module and then to make some kind of aparatus that injects current into the lid and sides to heat it up with high current power.

I thought maybe to wrap RF-braid around the part to act as one of the electrodes, tensioning it some how, but I am at a loss how to get the current into the lid, do i need multiple contact points to inject current around the perimeter ?

The other idea I have is to glue a handle to the top like the previous case, smear around it with antiseize then drop a block I heated on a hot plate on top of it (the block would have a hole in the middle around where the pull handle is), then immediately start pulling on it so it melts the solder.

It is like a thick aluminum gold plated can that they some how soldered a lid to, diameter 1.5x1.5 inches, similar to what is in the 8447 amplifiers from hP.

I am curious what a good high temperature thermal transfer compound is, the best that comes to mind is antiseize, which may function at solder melting temperatures.

Or perhaps instead of heating it on a hot plate, to wire a the resistance transformer to the block to make a hot block that sits on top and is heated by the transformer (voltage is ~1V), but I am almost thinking externally heating it is better since its more similar to natural soldering?

The idea is not to damage the parts or damage the coaxial SMA connectors. I think it 50 ohm so they are delerin, if it was 75 ohm I would be much more confident with heat.. usually when I have a thermal problem I switch to 75 ohm teflon connectors, which fare far better.

Or I thought to put a bunch of heat block on the connectors (I have a heat block 'putty' thing for welding) and just blast the top with a torch.

Or try to preheat the whole thing to a safe temperature on a hot plate then use a pace electric solder pump to try to get it off?

Or to heat the block up first but leave an electrical connection and then heat the block more right after putting it down to try to maintain temperature better.

[Kean]

Sounds like you are way overcomplicating this, or I am missing something...  Just use a hot plate?

[mikeselectricstuff]

A Metcal or JBC iron with a big fat tip, combined with desolder braid will open most RF cans IME

[coppercone2]

I am worried a hot plate will give too much heat into the thing. I suppose I can put a hot plate on it upside down. I don’t want to risk desoldering an 4 pin tiny rf package by heating it upside down and as for heating it right side up I am worried the 50 ohm coaxial connector dielectric will get damaged since the can has high thermal mass and it will take a long time to cool. Maybe a upside down plate would work.. maybe solder braid would weaken it enough to pry open. Maybe that’s ok since this one is 3ghz only... I never had solder wick work good enough for a part to fall off, usually it just gets weakened. I don’t want to damage the housing

[coppercone2]

This one looks like a thick can, like the ones that are normally screwed on tight. Not a wimpy sheet metal one over a pcb

[Tom45]

How about cut it open and then solder it back together to close it up.

[coppercone2]

I don't know how I would do that without a milling machine with a very thin slitting saw. That is a extremely nasty solution if you use the things that most people have available IMO. I also suspect its plated with something that makes it solderable but once its cut your not gonna easily solder it back together (a fucking fiasco with aluminum soldering), I don't think its solid copper, but maybe.

I was wondering that they use to make these perfect RF cans, without solder spillage and a solid plating. I assumed it was some kind of resistance heating and solder preforms. I think this is like a few MM thick. Everything else I saw similar looks like for the same form factor it would use 4-8 screws

It looks like this thing but without screws
https://www.ebay.com/itm/Mini-Circuits-ZFL-500HLN-Low-Noise-Amplifier-10-to-500-MHz-50-Ohms-SMA/124555083878?epid=1001836394&hash=item1d000fc066:g:YFwAAOSwsTxgGFF3

[geggi1]

Hot air is fantastic on shield covers. Use a hotair-gun with a nossle.

[coppercone2]

alright I am gonna try to take apart the device again and measure the module cover thicknes. I remember a while back I looked at it and thought 'no way' but based on how confident everyone here is, i think I need to take another look, I just recall feeling horror when I saw this thing compared to other shield cans in regards to thermal mass. i recall it looking like a bunker compared to some shantys. shield brick seemed more accurate (way more then a can) but maybe I am mistaken and it still has a very thin lid in regards to the body.

keep in mind though based on how good they are, you can have real beast shield 'cans' that are 100% welded steel (yes, because it is low porosity, they just use alot of steel instead of supposedly more conductive materials in some gear, due to the difficulty of sealing), just they are not common. I was under the impression that the lid is like 3/16 or more thick but I will make sure it is that outlandish

[esepecesito]

I am worried a hot plate will give too much heat into the thing. I suppose I can put a hot plate on it upside down. I don’t want to risk desoldering an 4 pin tiny rf package by heating it upside down and as for heating it right side up I am worried the 50 ohm coaxial connector dielectric will get damaged since the can has high thermal mass and it will take a long time to cool. Maybe a upside down plate would work.. maybe solder braid would weaken it enough to pry open. Maybe that’s ok since this one is 3ghz only... I never had solder wick work good enough for a part to fall off, usually it just gets weakened. I don’t want to damage the housing

I think whatever you do, the metal will get to the temperature of liquid solder, no matter what you do... Metal conducts very fast the heat... When they soldered it, it was already at that temperature.

[coppercone2]

need to a fix a few more things before I get to this job

[Bud]

Dremel around the perimeter.

[T3sl4co1l]

I would set up an induction heater, but, well, of course I would...

It's... kinda not too hard, if you only need a hundred watts or a few.  Wire up an inverter ("ZVS" Baxandall oscillator is a popular and effective option) to a resonant tank.  Just need enough, and good enough, capacitors to resonate with the coil, without melting themselves.  Or with fixed drive, a signal generator into a half/H-bridge (via gate drivers) is doable with care.  Coil should be litz cable or copper tube (water cooling is probably necessary here, even at this power level), but if localized heating is desired (or doesn't matter), an open-face E core will do a fine job.

Tim

[coppercone2]

I have a powerful 2000A variable tap resistance brazing/soldering transformer

can you do a 60Hz induction heater with this? I could get a 20A variac to tune it

I assume it would cause some kind of problem but I kind of want to develop that resistance heating system further, I only use it for panel beating right now. It is a proper transformer meant for short circuit operation

I do not want to deal with ZVS

anyway 60hz induction would still be interesting even if its not suitable for this application, i am sure that is a good post, i feel like its really rugged and has a high safety and reliability factor