Heat Trace Spill Tray Design Rules of Thumb?

[Rosalynn]

I need to heat trace a spill collection device to prevent icing up in winter.
The heat medium is glycol/water; I'll find out the flow rate and supply temperature next week.
The tracing will likely be laid in 2" of light-weight concrete, with the spill tray (steel) laid on top of that.
My hokey first guess design is to lay the tracing (1/2" tube) back and forth in 6" intervals the length of the tray. About 56 meters (184') of tubing would be required to do this.
I don't think that, after 56 meters, the glycol will be very warm on a cold day. I could divide the tubing into two 28-meter segments, with their own supply and return connections. Or 3 sections, or 4.... Does anybody have a shortcut calculation method (or experience) to help decide how many tracers? I could do this from first principles, but why reinvent the wheel.
Alternatively, does anybody have info on what the losses to the ground beneath would be? My site can go to -40C (-40F) in winter for extended periods (weeks!). Wind is not a major factor. I want to melt water that enters the spill tray so that it stays empty (environmental requirement).
The spill tray is 6ft x 12ft and is shallow--average depth 1 foot.
Also--does anybody know of insulation that might be used beneath the slab?
Thanks,
Rosalynn

 

[isofilm]

Hi Rosalynn,

I do know that you will want to insulate the slab from the earth as well as reasonably possible. You might start looking at burial grade closed cell glass foam, such as FOAMGLAS from Pittsburgh Corning.

Your glycol source PQ curve and temperature will be needed to finish the solution. Also, the geographical location of the spill tray may be helpful.




Best regards,
Daniel L. Thomas, CTO
Novel Concepts, Inc.

http://www.novelconceptsinc.com

 

[Rosalynn]

Thanks isofilm, I'll check into FOAMGLAS. Hadn't heard of it before.

Sorry, what's a PQ curve? I'm not familiar with that term.

I believe that the glycol supply temperature is 70C, but need to confirm this (it may be a bit hotter); the geographical location is northern Alberta.

Regards,
Rosalynn

 

[isofilm]

Hi Rosalynn,

A PQ curve is a pressure vs. flow curve.



Best regards,
Daniel L. Thomas, CTO
Novel Concepts, Inc.

http://www.novelconceptsinc.com

 

[Rosalynn]

D'oh! Sounds like what's called a pump curve around here.
Ok, I'm going to the site this coming week. Will get the PQ curve and report the pertinent details asap.
Thanks,
R

 

[Rosalynn]

This is info for Isofilm, regarding the PQ curve for the glycol pump.
The tracing for the spill tray will be part of a larger heat tracing network in the plant. The pumps deadhead at 350 ft of TDH and pump from a suction tank kept at 11 psig. The operating point on the curve is at 1633 USGPM and 375 ft of TDH, while the "runout point" is at about 2600 USGPM and 225 ft TDH.
My interpretation is that, as long as I don't draw too much glycol flow, the spill tray tracing has about 30 psi of pressure drop to work with (the glycol flows through quite a bit of piping and a few exchangers before it gets to our tie point, and there's a few hundred feet of line to get back to the pumps as well).
I'm thinking that, given the variability of the weather in the area, a 1/2" globe valve on the tracing outlet might be good in this service. If the truck drivers see ice or tar in the tray, they can open the valve a couple of turns to melt it. If things seem too warm (the supply is actually at 220F(!)), they can pinch back on the flow.
I'd be interested to know what you think of this. Would you use an orifice plate as well?
Thanks,
Rosalynn