Manifold design

[griff500]

I want to make a manifold to run UV and a media reactor for GFO.

Both would have ball valves to control flow and a flow sensor on the UV and on the main return, with the return pump speed controlled by 0-10v to maintain the flow back to the tank regardless of flow to the other equipment.

I was thinking about having a pump in the RO section that connected to the media reactor using a tee valve so that when new media was put in the reactor I could turn the tee valve to direct RO through the reactor instead of water from the tank and flush that out into a bucket using a detachable hose from the outlet. Perhaps that pump could handle duties for ATO as well, so the tee valve to the reactor for RO would have two positions - one to the reactor and the other to the sump (although the ATO flow rate would be a bit higher than would usually be the case so it would need to have a very short run time and I'd have to account for the soft start and getting RO through the pipework and into the sump, so maybe just a pump for the reactor flush and let my GHL Maxi Doser take care of ATO duties as currently intended).

Not necessary of course, but perhaps neater than needing to have a bucket of RO on the floor in my utility room, and extension lead for a pump as I don't have a socket in that room and GFO flavoured RO down the kitchen sink isn't something that makes my girlfriend particularly happy.

Also, I use a D-D 20W UV unit. If the pipework was supported well then could the UV unit be too heavy to simply hang freely while connected to the pipes or would it still need to be attached to the cabinet? The intention for the media reactor will be for it to hang above the sump (inspired by @Broadfield ) with the inlet hanging straight down from the top and entering directly into the reactor vertically, thus saving sump space.

 

[Flippers4pups]

Can you draft your design out on paper and post it?

 

[scriptmonkey]

I used ball valves on things that was a simple on off... I used gated valves on things that needed to precise control of flow. I screwed up on one valve, the Drain into the sump, I made it a ball valve and in reality I wish it was a gate, that way I could control the height a little bit easier in the overflow. So.. if you want to do control other than on\off...use a gate in my experience.

 

[Flippers4pups]

Put unions on everything so you can replace valves. If you don’t, you’ll wish you had.

 

[scriptmonkey]

Put unions on everything so you can replace valves. If you don’t, you’ll wish you had.

OH!! That too!

 

[griff500]

Can you draft your design out on paper and post it?

I'll try and knock something together but it won't be pretty.

I've got to get the tank in place to see where the UV unit will fit but essentially the return pump is in the back of the sump so I'm thinking of going straight up to the top of the cabinet and along to the right, dropping down and looping back into the return on the right. From the section going along the top of the cabinet, where it will be clamped in place, I'll have tee joints going down and hang a reactor off one and figure out the UV once I've seen things.

I'm tempted to route it to the back wall of the cabinet but I'll have cable trunking at the top of the back panel, although I could go up the back and then use a couple of 45-degree joints to get the pipe around the trunking to the top of the cabinet. That could work nicely.

I'm tempted to use clear PVC pipe in the white cabinet - it could like nice, although I'm not sure how long it would look nice for...

 

[griff500]

Something like this?

The outlet from the reactor could be removed and replaced with tubing into a bucket when flushing or just have it as tubing all the time but a little longer than required so it would work for both purposes.

To flush fresh media it would simply be a case of redirecting the ATO flow up using the T valve and using the L valve to send RO to the reactor instead of water from the return. The return pump would be controlled by 0-10v automatically based on the last flow sensor in the chain.

The pump used for the ATO/flush would have to be powerful enough for the flush but not too much for the ATO. I could control it using 0-10v and then have a Maintenance task for the flush that increased the flow rate used and a lower flow rate when it's just performing ATO duties.

Thoughts?

 

[griff500]

The 45-degree bits at the top would be to allow for the cable trunking.

If the pipe running along the top ended up in the right position then the reactor could hang straight down instead of running further along the top of the cabinet.

Oh. and there will be a gate valve to control flow through the reactor.

Those T and L valves are certainly not cheap and the same result could be achieved with a number of normal ball valves for less money, albeit not quite as neat perhaps.

Hopefully that's making sense.