I'm rebuilding an engine hatch/sundeck.
The bottom side is finished gel coat. The top gets a separate upholstered panel.
The hatch itself was laid up with a 3/8 plywood core and covered with just a layer of mat/resin.
Mostly rotten and wet. Stripped it out and laid in end grain balsa set in hull and deck putty. Laid a layer of 1.5oz mat over that.
Set it up with about a 1/2" crown like it was originally..
Came out great, two weeks later it's crowned in the other direction. Curling up like a potato chip.
Currently it's stripped back out again to get fixed, but before I try again I'm wondering where I went wrong the first time..
Any ideas?

Sounds like way too little fiberglass.

We had a new from scratch engine hatch made for the Mag when changing it from the walk around lift up engine box to the large beam to beam sun pad. I guess I could have done it myself if I had the facilities and I would have used a foam core but the guys doing it used balsa core . It must have been at least 3/4 inch balsa as the finished product is a little over an inch thick and it's about 4 feet by 6 feet .
The balsa was well saturated with resin on both sides before any cloth went on it (no putties used) and the thing is so rigid you could use it as a dance floor for 6 people. However , it's heavy as fox . Not a fraction of an inch change in it's shape since day one. Maybe you have an issue of not enough resin being used in the initial layup ?

Very well could be madbouz. The bottom side going in was just laid into the hull and deck putty which I did add resin to to thin it out some.
That was spread with a 1/8" V-notch trowel.

watch your catylist ratios and also use a high quality VE resin at the least. Loosing control of a laminates shape can be caused by to fast a cure out or low cost GP poly resins. If you are trying to maintain a specific curve you want the cure to be slow and on the cool side.

over use of resin and cloth is also a major problem. These things should not weigh a ton, that tells me they are heavy and wet. An engine hatch does not need much more then two layers of 1708 on both sides of a half inch core. With a quality VE thats a pretty stout panel if its well balanced and void free.

I'll have to admit, I probably mixed a little hot..
I've been using PE resin, good quality from Merton's.
Also am guilty of laying it wet on wet, and not letting it cure between steps.
Will watch ratios closely this time, and rather that the 1.5oz mat over top going to do 3/4 oz , 2 layers with cure time in between.

I agree completely. When I first got my hands on my finished product I was surprised to put it mildly. I figured that if any production boats that had used balsa core construction were made like this hatch they'd need 5x's the power to just get on plane.
At least I know the thing will be around for a long time .

its a pretty common mistake and i have to admit in my earliest days i could not get it into my head that i did not have to actually see the reaction and feel the heat for it to be working (I'm talking early days working on golf car bodies as a teen lol) For PE 1% at 70* is it . . . no more. Steve Koss is the one that helped me understand how important cat ratio is. To hot bring all kinds of problems. Shrinkage, weakness, brittle, etc . . . .

The factory Twin Step top gun hatch i sell is around one hundred pounds the OEM hatch on a 01 and up gun is about two hundred and forty pounds

Pretty amazing . How can they justify that when you can do it so much lighter ?