I agree that the EJMA diagram proposed will eliminate pressure thrust amongst most EJs, provided the EJs are identical and there's no pressure isolation device in between.

I don't envision running into problems unless you're altering an existing failing design, and you're forced to make things "better," instead of "per new design rules."

I'll also point out that the expansion joints will probably be activated while the expansion joints are depressurized. You can expect to see fatigue to set in more readily in the EJs in this configuration if it's not reasonably designed.

I'm not going to jump to conclusions as to what you think a given application *should* be, so instead I'll just genericize.

Let's talk about some configurations.
1) Straight line with 2 EJs.

You'll end up with double the pressure thrust, but the effective stiffness will be halved compared to a single expansion joint.

The benefit is that you practically eliminate forces being transmitted laterally through the pipe, perpendicular to the hinge pin. This is compared to a single expansion joint which permits free rotation through the pin, but the pin still transmits load.

The detriment is that you have a rotating section of pipe between 2 EJs. If there's significant rotation, the two outermost flange connections of the two expansion joints get pulled closer to each other.

I don't think CAESAR understands that, but I don't have a way to test it at the moment. So, buyer beware.

2) 2 EJs that are separated by one 90 degree turn.

So I don't think there are many pros to this configuration, and a lot of cons.

The pressure thrusts are mutually reduced. That's about the only pro I can offer.

However:
If your pipe is installed in the horizontal, and your hinge pins in the vertical, you're now supporting the weight of your pipe on those two hinges. Your EJ manufacturer will thank you for your repeat purchase if you don't adequately support this. And by virtue of supporting it, you're reducing the effectiveness of the EJs.

If, for some unknown reason that you install the pins in the horizontal, your expansion joints will activate based on weight alone, and the elbow simply falls and creates a pocket. Same deal - support needed, and the support reduces EJ effectiveness.

Thus you have vertical configurations. Hinges should probably be parallel to the ground, but the weight of the pipe is likely going to extend the expansion joint in the vertical, thus creating a low point. This may or may not be tolerable.

3) 2 EJs with 2 90s in between.
There's not much to say about this that hasn't been discussed before. Take a look at your own diagram and note the substantially large displacement

4) Combination - 1 + 2
This is a configuration that's prescribed for hypan ball joints in systems with large axial displacements. This allows the pipe to act more like a mechanism than a pipe. Note that CAESAR might not give you an accurate answer, here.

5) Combination - 1 + 3
Take what your image says - note the deflection of the vertical run of pipe. That has a set amount of overhead loading associated with that displacement. The second expansion joint below the one on the riser helps take some of that out.

6) Combination 2 + 3
I can't really imagine a need for this, and nothing but headaches. Don't do this is my only advice.