I went through the same process a few years ago, and I will be receiving a masters in physics this May. I started school at age 35, so similar to you.
1. As Locrian and TMFKAN64 said, you are going to take a huge financial hit unless you can find a company to pay for school *and* pay you a good salary. Maybe you can do it, but I think it's pretty rare. If you take the more common path of doing an RA/TA, the financial consequences will be serious, as you will have a huge (though hopefully temporary) pay cut. You're first job out of school may also pay less than your current job. You really need to think about this aspect of it.
2. You didn't mention a spouse or children - if you are married, you better make sure your spouse is 100% supportive. Mine is, and that has been an enormous help. If you have kids, or are planning on kids, the financial part becomes even harder - children are expensive. The reason I'm exiting with a masters is that the cost of children (which were not entirely planned when I started grad school) proved to be too great.
3. You don't need a physics bachelors to get into physics grad school. My undergrad was engineering, my career was 12 years as a computer programmer. I had freshman & sophmore physics from undergrad. Prior to applying to grad school I took three additional upper level undergrad physics courses.
My attitude was that I needed recommendation letters from physics professors, so my main purpose in taking the classes was to impress the teachers and get good letters. Of course it was good to take formal classes also (more efficient than self-study in my opinion), but I felt the letters were going to be very important for getting accepted.
4. You should probably take a sample physics GRE test to see where you are. Be warned that the physics GRE is *hard*, so expect this and don't be freaked out the first time you take it.
You will need to take general GRE too, but I wouldn't worry about that too much. I did better taking the general GRE in my thirties than I did when I took it in college!
5. Your work experience - especially with an engineering background - will likely be viewed as *very* valuable by experimenters, and you should have no problem getting an RA starting in your first semester, if you make an effort to find one. Do not underestimate this advantage you have over other students! You will be considerably more mature and self-directed than all or nearly all of the other grad students, and this will be a big help.
6. It wasn't clear from your post whether you are a programmer, but programming skills are also valuable. This might be a good skill to leverage if theory interests you - there are areas of theory where simulations are common.
7. Theory vs. experiment. Many/most want to do theory when they start, but of course end up in experiment. I was basically the same. What is your goal is doing this? Do you want to become a physics professor (very long and challenging road, especially starting in your mid-thirties) or do you just want to find a more satisfying job? If the latter, experiment is probably going to put you in a better position.
8. Think about going to a school with connections to national labs - should help with jobs after you are done.
9. I wouldn't worry too much about what particular area of physics you want to specialize in. I think that nearly any area is very interesting once you start to learn about it.
10. Finally, it's odd to be an older student in grad school. You are at a very different place in your life than the other students, especially if you are married. You may be the same age as some of your professors! You will be the junior person in your group when you start, possibly taking direction from a senior grad student 10 years younger than you; check your ego at the door.
Good luck! It's been a tremendous experience for me, and despite the financial difficulties I'm very happy I did it.
