'calibrating' the DUT instrument connections

[BillA]

as has been previously discussed by many,
the measurement connections to the DUT have their own 'losses' and these need to be quantified in order that the gross readings may be corrected to arrive at a net value for the particular DUT parameter

some preliminary data was described in the thread on "C", this has been redone and expanded
for these tests the inst crosses, RTD ends, and connecting tubing were insulated with 3/8" closed cell neoprene

the hydraulic data is straightforward and the effect of temperature can be seen,
though not really in the flow range of significance - below 1.5 gpm - with the instrumentation used

of greater interest to me was the temperature data taken at the same time:
the ambient temp was held at 25°C and the dT measured at 11.5, 5.5, and 0°C temperature difference; then run again at 15.3°C with 0°C difference

there is a temp change w/o any DUT, for higher accuracy this must be addressed

the WCing testing community might consider the use of these (cheap) crosses as a de-facto std, everyone would have the same corrections
(there is an inst offset included in these curves which could be backed out)

[bigben2k]

Thanks for that.

I proposed (here) filling up the cross to have straight, uninterupted flow, similar to a straight pipe.

Here's a (partial) diagram:

[BillA]

use a tee, easier
ftgs w/0-ring from McMaster-Carr

Ben, if anything different is done to the crosses new cal curves are required
if you, or anyone else, makes the crosses as described then these curves could be used

not many will have the capability of generating these curves, why they are posted

[bigben2k]

Yep, I realize that (re: cal curves).

So... (schedule 40) 3/4 PVC tee, then 3/4" to 1/2" adapters (glued), then 3/8" ID tubing and 6" Silicone tubing.

I like that setup.

We're still going to have some variations, unless you can detail the setup here (i.e. protrusion of the pressure port inside the tee, protrusion of the temp probe and at what depth). Am I just being picky?

When you state 6" OAL connection tubing, what precisely does that mean? Does that include the barb overlap?

Bill, you picked this arrangement, but I can live with a schedule 40 tee, with a tapped pressure port as you have (I'm assuming it doesn't protrude inside the tee?). I would have proposed schedule 80 parts, which are threaded.

Should we go into ID variations of barbs, or agree on a single spec and/or supplier? Or ignore as irrelevant?

[BillA]

I'll gin up a BOM tomorrow, off the shelf stuff/McMasters-Carr
the previous version was sch80, too small a bore
pressure tap/ftg mtd opp to branch, flush to bore
barbs are CRUCIAL, use Eldon-James 1/2"NPT x 1/2" single barb, no mods
connection tubing is 3"OAL, 3/8"ID pushed over the 1/2" barb (each side of DUT)
all insulated, I'll take a pic later

[bigben2k]

Many thanks!

[Les]

Am I right in saying that,now, at 25c Environment Ch and 25c Coolant- Entry- T-joint there is a ~ 0.02drop in temp across the "6Tube +bits"'s DUT.
Previously (without? neoprene(inter-alia?) was ~0.06c drop

Amongst previous data(link) making some sense of Series1
Which had taken to be 0.52"ID?(cf 0.93"ID? of 3/4"Ts)
Crudely analysing in terms of Ev=(1-a)V^2/2gJCp and Ec=(Tw-Tx)/cosh[Rx*sqrt(h*Ap/kAp)]

Edit: All crap.These are the errors which should cancel if using eqi-sized "in" and "out" T-things

Have you abandoned smaller Ts?
Thought had possibilities due to better mixing(Re ^) and if equalized to 3/8"ID of tubing,possible, less disturbance of flow.

Apogee'ed brain needs time to re-acclimatize.
Possibly no difference, possibly it is jumping out of the frying-pan into the fire.
Probably best, for own sanity, to take hint from others and jump out of both..

[Incoherent]

Originally Posted by unregistered

the hydraulic data is straightforward and the effect of temperature can be seen,
though not really in the flow range of significance - below 1.5 gpm - with the instrumentation used

of greater interest to me was the temperature data taken at the same time:
the ambient temp was held at 25°C and the dT measured at 11.5, 5.5, and 0°C temperature difference; then run again at 15.3°C with 0°C difference

there is a temp change w/o any DUT, for higher accuracy this must be addressed

the WCing testing community might consider the use of these (cheap) crosses as a de-facto std, everyone would have the same corrections
(there is an inst offset included in these curves which could be backed out)

Bill I need to get this straight, having difficulty re-engaging brain. I don't really understand.

Where is the water temperature measured?
The correction would be for the RTD in the "T" piece? I guess I am wondering which is right...

The hydraulic behaviour is I think real (that said without any deep investigation at all) and therefore requires no correction, just a statement of absolute water temp. ?

I am not at all up to speed here.

[Les]

Originally Posted by unregistered

...
of greater interest to me was the temperature data taken at the same time:
the ambient temp was held at 25°C and the dT measured at 11.5, 5.5, and 0°C temperature difference; .....

You held ambient at 25c then decreased coolant temperature to create dTs of 11.5, 5.5, and 0°C ?

This revealed apparent Dissipation by coolant at dT=0 and Absorbtion at dT=5.5 and dT=11.5 ?
As shown in attachment?

Excellent and I like* it if I have + and - signs correct.

* Not sure,yet, about forcing me to use formulas

[BillA]

Les
yes, but there may be an offset - I need to think for a bit on it
may be ok ??
thinking of the swapped offset described in "C"

Incoherent
the instrumentation 'cross' is actually a modified tee, to function as a cross, with the branches having a RTD port and a pressure tap
the crosses are on either side of the DUT and are shown bare in the first post and as now insulated below

the measured difference is a steady state avg (peak to peak / 2) between the inlet and outlet
(from the RTDs within the crosses, each corrected per its cal on that inst)
equip kinda pushed to get these values
want the spreadsheet ?

EDIT
the temps are NOT a simultaneous difference but independent max-mins,
the lag varies with velocity
lag

[BillA]

well, there was an offset in the temp data; the coolant RTD corrections were reversed !
data is a lot more sensible now

spreadsheets sent to those interested

[bigben2k]

Originally Posted by unregistered

...
spreadsheets sent to those interested

Please cc ben at wbta.us

[Les]

Just in time,about to look at 25c vs 15.6c.
Will still do,hopefully before Friday pub, but eases considerably.

I note:
considerable scatter at 25c(R^2=0.073), worse? at test coolant 30 or 35c, but doubt is relevant
level back at no-neoprene level, consistent with my thoughts on still air vs closed cell neoprene

Ta for spreadsheet

[BillA]

could re-run but doubt sig chg

ok so far, but the air/liquid cal ? is unanswered
(in my mind, all caled in flowing water for now)

sent Ben

[Les]

Originally Posted by unregistered

1)could re-run but doubt sig chg

2)ok so far, but the air/liquid cal ? is unanswered
3)(in my mind, all caled in flowing water for now)

sent Ben

1)Agree, not worth effort
2)It is not - looking there but in terms of W
3) Also happy.

[Les]

Think, you are now set up and calibrated to test the "6" of Tube" with any fan you wish.
Initial thoughts say you have to repeat the procedure with each radiator(with auxiliaries attached).
Hope I am wrong

[BillA]

yea, needs to be redone
agree that expressing in W is more understandable for most (but

[BillA]

.

[Les]

Originally Posted by Les

...hopefully before Friday pub,...It is not - looking there but in terms of W...

Musings attached