Hey Jeff,<div><br></div><div>I changed the time step and tolerances and still see the same trend. Here is what I changed them to. The simulation took about 5 times longer.</div><div><br></div><div>Time Step: 0.01 [Hour]</div>
<div>Tolerance Integration: 0.000001 [-]</div><div>Tolerance Convergence: 0.000001 [-]</div><div><br></div><div>Also so you know, I attached the TRNSYS schematic of what I ran. A very simple system for testing as you can see. I set a constant flowrate of 5 gpm on the load side of the HX that never changed, and I varied the source side from 3 gpm to 12 gpm over the course of 6 trials. The HX is also 100% effective for testing purposes, but the same trend occurs with any effectiveness under 1.</div>
<div><br></div><div>I attached a pdf of the results if you want to take a quick look. The numbers from the default tolerance test and tighter tolerance test are virtually the same, and show that same trend: the performance of the collector and the heat transfer in the HX increase from 3 to 5 gpm, and decrease from 5 gpm and up.</div>
<div><br></div><div>This does not happen when do not use a HX.</div><div><br>Is this what the trend should actually be or is TRNSYS (or me) doing something weird?</div><div><br></div><div>Thanks for any help Jeff. I appreciate it.<br>
<br><div class="gmail_quote">On Thu, May 19, 2011 at 8:57 PM, Nicholas LaHam <span dir="ltr"><<a href="mailto:ndlaham@gmail.com">ndlaham@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div style="word-wrap:break-word">I'll try that tomorrow Jeff.<div><br></div><div>I am using a constant effectiveness and I am leaving pipes out of the simulation for now to figure out what is going on.</div><div><br>
</div><div>Should this be the trend?</div><div><br></div><div>Thanks,</div><div>Nick</div><div><div></div><div class="h5"><div><br><div><div>On May 19, 2011, at 2:11 PM, Jeff Thornton wrote:</div><br><blockquote type="cite">
Tighten your tolerances way up and shorten your timestep dramatically and see if the trend changes. I'm also guessing that you're using a constant effectiveness HX? Are you using pipes for capacitance or just a collector and HX?<br>
<br>Jeff<br><br>Jeff Thornton <br> President - TESS, LLC <br> <br> 22 North Carroll Street - Suite 370<br> Madison WI 53703 USA <br> <br> Phone: <a href="tel:608-274-2577" value="+16082742577" target="_blank">608-274-2577</a> <br>
Fax: <a href="tel:608-278-1475" value="+16082781475" target="_blank">608-278-1475</a> <br> E-mail: <a href="mailto:thornton@tess-inc.com" target="_blank">thornton@tess-inc.com</a> <br> Web: <a href="http://www.tess-inc.com" target="_blank">www.tess-inc.com</a><br>
<br> <br> <br> <div style="background-repeat:repeat-x"><strong>----- Original Message -----</strong><br><strong>From:</strong> "Nicholas LaHam" <<a href="mailto:ndlaham@gmail.com" target="_blank">ndlaham@gmail.com</a>> <br>
<strong>Sent:</strong> Tue, May 17, 2011 10:55<br><strong>Subject:</strong> [TRNSYS-users] Type553 Unglazed Collector Theory<br><br><table width="100%" cellspacing="5"><tbody><tr><td><div> Hey all,<div> </div><div>I have a question about the Type553 Unglazed collector and how it's performance interacts when changing flowrates in two different scenarios. I was wondering if someone could help explain the trend in a practical way for me so I can explain it to my colleague.</div>
<div> </div><div>First off, when just running a constant flowrate and constant incoming temperature as the inputs to the collector, I can see that increasing that flowrate increases the useful energy out of the collector. This makes sense and is intuitive.</div>
<div><br>Now when I use a loop instead for the same collector, (say with a pump operating at 1 GPM just as a starting point), and a perfect heat exchanger with the same incoming water temp now on the load side of the HX, I see a different trend I'm having trouble explaining to my colleague.</div>
<div> </div><div>What is happening is that the useful energy increases like the first scenario, but only up to the flowrate that is on the load side of the HX. <i>ex. So if I have 5 GPM on my load side, I see an increase in performance up to 5 gpm on my collector side, and then the performance begins to drop after increasing it any further.</i></div>
<div> </div><div>Another way to say it is that whenever my flowrate of the collector array is greater then my load I have weakened performance. It seems that a 1:1 flow ratio on each side of the HX is the optimal method (assuming same specific heats).</div>
<div> </div><div>Can anyway explain in a conceptual manner why there is this decrease only when using the HX? I was thinking that because the load side of the HX is the limiting flow (the one that isn't changing) then because the collector flow is higher, that not all the heat from the collector can be transfered, leaving the outlet of the HX on the collector side hotter than if the flows were balanced. Thus in turn lowering the efficiency and available energy to be collected the next time it passes through the collector.</div>
<div> </div><div>I can actually see the source side temp coming out of the HX hotter than the incoming water when the collector flow is higher, but my colleague still thinks the energy should not drop. He believes that when increasing the flow up to a very large number over many trials, the useful energy will always rise up until it levels off asymptotically, whereas in my case, it is doing more of a bell curve and dropping off after it becomes higher than the load flowrate.</div>
<div> </div><div>Thanks for any help.</div><div>Nick LaHam</div> <hr size="1" noshade><pre>_______________________________________________
TRNSYS-users mailing list
<a>TRNSYS-users@cae.wisc.edu</a>
<a href="https://mailman.cae.wisc.edu/listinfo/trnsys-users" target="_blank">https://mailman.cae.wisc.edu/listinfo/trnsys-users</a>
</pre></div></td></tr> </tbody></table></div></blockquote></div><br></div></div></div></div></blockquote></div><br></div>