Hydrodynamic Bearing/Water Column

Not enough to go on.

Reasons you can't get a good seal?

Xducer drawing?

Picture?

There is a slot on the xducer cover that is probably part of the assembly process to allow clearance of the BNC connector. This is where the leak is occurring, Silicone just does not contain the pressure.

Using silicone as a dam rarely works well.

I like to use a product called Threebond 1104

I modify the application regime by coating both surfaces with a thin film first and allow it to skin before applying more and assembling the parts.

Orient the work so that the sealant "pools" favourably around the joint if you can.

Let it set for a day before subjecting it to pressure.

I have also found that tiny leaks sometimes heal themselves if just left alone.

To return to your original question:

If you are going to try an alternative to the silicone remember that nothing adheres to silicone - not even more silicone. And silicone, once applied, is extremely persistant. I would use wet and dry sandpaper with a fresh piece each circuit of the device. Use acetone instead of water with the sandpaper.

Not enough to go on.

Why open mouth with water pressure. Cavitation?

Hospital ultrasound works with a bag and gel outside with no pressure.

Why is that not good enough for you?

That is for starters.

The slabs have a good amount of scale (Fe3O4) that is tightly adhering to the surface. This method rides over the scale and all loose scale is just swept off the slab. Couplant on this scale (many square yards) is expensive. I am limited to in plant water pressure , 35 psi.

Cavitation? Yep.

I would try 3M 5200. Readily available (Home Depot in the U.S.) High adhesion strength (bonded to fiberglass it will peel the gelcoat off if you try to pry off something glued to it - personal experience) The downside is the long cure time. You would need to give it at least 3 days to cure.

Bulkhead connector? That'll give you more sealing surface?

Could you use a thin film of plastic to seal the tube and place the xducer on top of that?

It seems I am obtuse, or your explanations are still way off the mark. Whatever.

There is a principle in engineering, that there is more than one way to skin a cat.

Then there is inverting the question. Then there is thinking outside of the box.

One example:

Surface is rough. Cleaning it is bothersome. Fine. The purpose of the exercise is to ensure the ultrasound waves get coupled into the substrate reliably. Put a heavy polyethylene or mylar foil down, and feed it with water at its approximate center. That is, between the sheet and the concrete. That will float the foil, give perfect coupling, and allow a closed (bag, hospital type) ultrasound type transducer to glide over. The transducer is decoupled from the water pressure requirement. You may or may not have to maintain a small water or gel coupling between the generator / sensor and the foil, to make it sure, all ultrasound energy has a clear liquid path.

Problem solved.

A little complicated and time consuming. If there was just one slab then fine, but incoming can go into >500. This unit will finally be semi automated where there is a raster scan and reporting by C scan. The entire action is to check the slabs for center line shrinkage as the supplier did not perform this inspection and i have setting outside my office around 50 rolled plates that are all reject. One has an obvious hydrogen bubble in the center that we opened up and literally looked into the center of the plate. This is a steel mill and asking inspectors to do a delicate operation is like asking for world peace.

Picture of the bubble to follow.

I sense your frustration, and I symphathise, for what it is worth.

But, I and my compatriots are lacking explanations / informations your last note partially revealed. Keep telling us. You are working in rough environment indeed, making it difficult. Dumping info on us helps.

Is the leak from the fitting to the red hose? Where is the leak exactly in your OP picture? Are you trying to cause cavitation in the transducer, or is that unwelcome? What is the frequency (2.2 megahertz sounds way too high, 2.2 kHz would be ultrasound)? Do you have to fill the transducer with water, or could a more viscous fluid (mercury?) be used, which would not get through the seal?

Is the leak from the fitting to the red hose?
Nope the red line is going to the scope, that leak is the weep hole which allows air out .I really dont have a picture of the leak, but i will post soon.
2.2 megahertz sounds way too high, No that is the correct frequency any higher and the wavelength would be too long and we would miss indications.
could a more viscous fluid (mercury?) be used, which would not get through the seal?
No , water is the common method.

Wave length gets shorter as you increase the frequency, not longer.

Yes that is the point. it is a balance of penetration and detection of smaller reflectors.

Slab surface typical.

Mother slab typical.

Sorry lyn. Senior moment.

That's OK. I've tried to rewrite the laws of physics before, myself.

Where in the original picture you posted is the leak you are trying to seal? There evidently is a lot of water flowing out from the water column and over the surface -- is that to be sealed? I suppose you are trying to avoid cavitation -- am I right? On reconsideration, what is the frequency of the ultrasonic transducer that agitates the water column? Although the common method uses water, I see no reason other than availability that water and its dissolved gases should be preferred over an elastic fluid like mercury for stably connecting a slab surface to a horn.

This is what started all the ruckus. Our metals guy was walking near the heavy leveler and noticed this 4" plate. When the plate was lifted the bulge was on the bottom as well as the top. The picture with my boots in it is looking at the center of the plate, not another plate.

As you can see the hole has become an ash tray.