I was having coffee this morning and looking at numbers from this thread and the more recent calibration thread.
If you know someone with a 1020 bench, it would be IMO, worth your while to have one of your heads tested on it. Specifically, one that people claim your top numbers were high on.
If, when tested on the SF1020 bench, the numbers were closer......well, that's rather self explanatory now, isn't it
I've been thinking about this problem for a while and I was just wondering if the problem was some sort of turbulence in the cabinet issue would it be possible to do a test where you at higher lift keep the actual flow number the same percentage wise on the inclinded by purposely reduce the depression you're using for testing. Then correct those numbers back to your standard 28" or whatever you're using. It would be interesting to see what that does to the numbers.
You and Chad might do some comparative testing and see what happens. Test the same head on the same size bore adapter at 10, 25 & 28. Post the results.
Chad could flow it on his PTS unit as well....same criteria.
Last guy done with the head, ship it to Bruce for testing on his bench.
Something else that never gets brought up.........
How many guys are doing a leak test prior to testing a head and actually taking away those numbers from the test results? I'll wager not many, if any are?
I always record data at zero lift when I flow test and then that gets subtracted when I'm done because it's a good way to catch any screw up(leaving out the spark plug) before wasting time with a test that's not valid.
Most of the time it's a very low amount anyway if everything is ok.
Does your flowcom match your inclined when you do a leak test?
I've seen digital setups act funky when there's no flow but depression is applied in the cabinet.
I've only ever seen a couple of pics of the inside of a 1020 bench. That said I don't know how big the actual plenums are. I expect, based the look of the machine, they are a fair bit bigger than a 600's plenums are.
My hunch is the SF600 should have been upsized instead of using the 300 cabinet and coaxing more flow through it. I think that's 99% of the problem right there. I also think SF knows that and addressed it with the 1020 bench...
My Flowcom will match the inclined manometer very closely when checking on the 10cfm scale. Leakage for me is under 1cfm as long as everything is set up correctly.
I have noticed some Flow Com weirdness before if I don't allow it to warm up before testing on a cold day. If I come in and it's cold in the shop and I just turn on the bench and do a test once in a while the Flow Com will read zero when the inclinded is at about 15% and increase flow in either direction from that point. I can very slightly open the valve and get the Flow Com to show zero which is weird. If I just let it sit for an hour or so and then start testing everything works just fine. Most of the time I just leave the Flow Com turned on so this doesn't happen.
For reference; The PTS DM uses temperature compensating pressure sensors. Any reading at 0 on a PTS DM is related to how the averaging algorithm operates until it sees flow. It does not hurt to let it warm up though. Also for reference I use 1-5% electronic components on my circuit.
This may not be the same on the Flow Com it might not use temp compensating sensors nor 1-5% components? Using 1-5% parts does add to the overall cost of the unit and on the 5% parts it made a difference as the next jump up was 10% I felt the added cost was worth the end results.
Bruce
Who . . . me? I stayed at a Holiday in Express . . .
If I walk in, fire up the bench with a head on it, everything stone cold & run a test, then test the same port again after the bench is warmed up good, the numbers drop about 3 CFM. They will remain there though. I've gotten accustomed to letting the bench run with the head on it, one motor running with valve at max lift for several minutes, then do my testing.
If I do it that way, the numbers repeat. If I compare stone cold numbers to warmed up numbers, the cold numbers are always a few CFM higher. I don't know the specifics of how the averaging affects that. I haven't messed around with that feature since I calibrated the bench a few years ago.
I use a test plate on top of the test stand to verify the bench very frequently.
