Jean Pain Mound- PLEASE HELP - it's leaking!!

Hello-

I've just constructed a Jean Pain mound for heat recovery using polyethylene pipes with water running through. After construction I realized I had not used clamps, but had used two part pipe adhesive/ cement intended for pvc, not polyethylene and all joints but two were extremely tight- very difficult to force the couplings together in fact .
The two T-joints( with temp sensors for water inserted) were less tight but again I used the adhesive. (I am a student, this is my first project of this nature) After 3 1/2days the mound is already 175F. I tried to run some water though it and noted water leaking from the bottom, and only a very small amount made it through the loop.    

I do not have have machinery, this was all built by hand, approx. 22 cubic yards of material(wood chips/manure) vertically filled, supported by 60 bales of hay.The pipes are laid in coils on each layer of the compost mix.  I am trying to think fast since I know I have to reduce the temperature or risk crashing the mound before I even begin. I imagine the greatest weak spots are probably the two T-joints with the water sensors since they were the easiest to assemble with the least force. I am trying to think of a method of reaching those spots to make a repair safely....how to support the surrounding structure and create an opening large enough to reach these locations for repair, being mindful of the potential to get burned. That said, I realize these may not be where the leaks are. Also, of course other locations may have leaks, or develop them in time.

Is there any reasonable alternative to pulling the whole mound apart and beginning construction all over again?  My focus on this project this time around is precise data collection.

Advice and/or insights would be greatly appreciated.

This is one of those times when you have two choices:  You can a) spend a tremendous amount of time and energy trying to figure out a way around doing what you know needs to be done, or b) you can bite the bullet, pull it apart and do it correctly, chalking this up as a valuable learning experience.  I have found myself in this not-so-enviable position many times over the years, and almost without fail you will find that trying to find any easier way to do whatever it is that needs to be done will take more time, be more effort, and will not have as good results as just starting over.  

Can you do a test outside the sytem on joining two bits of new pipe with the PVC glue, see if it melts the pipes and fittings or if its simply not a tight seal. Maybe you can run some rad weld  or something in there to find the leaks and seal them up? Apart from that I'm strongly in favour of the bite the bullet advice in the previous post.

Hi Steve and Todd,

I really appreciate both of your replies. After careful consideration and input from a few people including my mentor I decided to at least attempt to reach the two t-joints I suspect could be the problem. If those are not it, or if I can't do this safely given the high temp of the mound it looks like it will be bite the bullet time. Since the mound is about 8' tall and involves enormous physical labor and time I felt this attempt is worth trying before starting all over. ( I used 60 bales of hay, not 30 as I mistakenly said before) I will keep you both posted.

Steve- the glue did not melt the pieces together when I used it- in fact I was able to separate one joint that I had glue about a minute or two after gluing.
But your suggestion inspired me to  test that kind of connector with *no glue*- I hooked it up and ran water through it. It did not leak at all; this gives me some hope that there's a chance it may only be limited to the one or both of the locations with T-joints I'm carefully trying to dig my way into.    Thanks  very much for your feedback.