|
Title: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: ddillenger on August 24, 2014, 11:32:45 AM It's cost prohibitive. The full panel of tests runs about 150 bucks per injector. If I do it myself I will send one 60lb, and one 47lb to get flow capabilities, offsets at varied voltages, and fow tested at 4 bar.
Ideally I'd like to send a larger sample as there's a chance we send one that is not representative of the group. Anyone want in on this? Regardless, the info will be posted here for everyone. Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: phila_dot on August 24, 2014, 12:10:09 PM Where are you sending them?
Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: ddillenger on August 24, 2014, 01:07:02 PM Deatschwerks.
Would like to go with their advance testing, get the linearity testing as well as the battery offset testing. http://www.deatschwerks.com/services/fuel-injector-services Thoughts? Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: phila_dot on August 24, 2014, 01:20:28 PM These are factory injectors with Bosch cal sheets right?
Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: ddillenger on August 24, 2014, 04:17:45 PM These are factory injectors with Bosch cal sheets right? The 47lb (550) datasheet has been disputed, and there are multiple incorrect sheets out there. The 55lb has no available spec sheet. Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: littco on August 25, 2014, 01:56:22 AM The 47lb (550) datasheet has been disputed, and there are multiple incorrect sheets out there. The 55lb has no available spec sheet. I'm sure I mentioned but I had 20 of the 298s flow tested . They all came back 600cc +/- 1% @ 3 bar.. no battery voltage testing done but we'd seem to have good results on the 550's data sheet.. It would be good some more accurate testing done though for the 298s Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: ddillenger on August 25, 2014, 02:12:39 AM Yeah, I'm not so interested in the flow numbers as I am the voltage offsets and linearity. After some FKKVS work lately, I am questioning the values we've been using.
Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: MIL_on on August 26, 2014, 05:34:43 AM You could also save your 150$ because this is exactly what i am planning to do with my set of 298s ;) Maybe i can even do it this week if i find some spare time.
of course i am going to share the data here. Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: nyet on August 26, 2014, 10:12:00 AM Flow testing is pointless at this point, like dd said.
What we need are offsets/latencies at operating pressure (in this case, 4 bar, not 3 bar). Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: MIL_on on August 27, 2014, 11:28:23 PM where did i say that i am only gonna measure flow rate?
i will measure dead time based on voltage. What i also use to do is to drive different duty cycles e.g. 5 and 10 ms and check how and if the dead time varies. Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: MIL_on on November 19, 2014, 03:07:10 AM some news on this:
i managed to test a set of the 298s! testing liquid: Gasoline with density 0,760 kg/l. Fuel pressure 4 bar (used FPR: 0280160575 ). 1. Qstat [g/min]: Code: Injector 1: 528.1 converted to cc/min: Code: Injector 1: 694,8684 deviation from average: Code: Injector 1: -1.08% Converting the 4 bar flowrate back to 3 bar means that the average Qstat of these INJ is 595.36 cc/min! After checking Qstat i checked the dead times @ 10ms pulsewidth from 9 to 15 Volt. Afterwards i checked the linearity behaviour over pulsewidth, but only @ 14 and 15 Volt (hope to repeat this another day with all Voltages, although i would say it doesnt matter that much, as you can just use it in FKKVS and not TVUB...). Yesterday we threw them i a 1.8T running GTX35 and tested them during normal conditions. I just used the deadtime for tvub i measured @ 10ms and had to alter FKKVS in low ranges of tevfakge to significantly <1 to meet low fr_w readings. Correlated quite nice with the table. But i also wouldnt consider that car to be the best choice if i want to verify anything. Greets Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: littco on November 19, 2014, 03:15:33 AM some news on this: i managed to test a set of the 298s! testing liquid: Gasoline with density 0,760 kg/l. Fuel pressure 4 bar (used FPR: 0280160575 ). 1. Qstat [g/min]: Code: Injector 1: 528.1 converted to cc/min: Code: Injector 1: 694,8684 deviation from average: Code: Injector 1: -1.08% Converting the 4 bar flowrate back to 3 bar means that the average Qstat of these INJ is 595.36 cc/min! After checking Qstat i checked the dead times @ 10ms pulsewidth from 9 to 15 Volt. Afterwards i checked the linearity behaviour over pulsewidth, but only @ 14 and 15 Volt (hope to repeat this another day with all Voltages, although i would say it doesnt matter that much, as you can just use it in FKKVS and not TVUB...). Yesterday we threw them i a 1.8T running GTX35 and tested them during normal conditions. I just used the deadtime for tvub i measured @ 10ms and had to alter FKKVS in low ranges of tevfakge to significantly <1 to meet low fr_w readings. Correlated quite nice with the table. But i also wouldnt consider that car to be the best choice if i want to verify anything. Greets That's very good news. I had 5 sets tested here in the uk and 598cc@3bar was the average. so good to see it correlates. I've happily been using the 550cc Tvub settings which looking at your pulsewidth table is a little different, so I will try that instead to see if its any better. Nice work though Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: MIL_on on November 19, 2014, 03:31:48 AM That's very good news. I had 5 sets tested here in the uk and 598cc@3bar was the average. so good to see it correlates. cool to see that different tests match up that nice, especially as i measured @4bar and with normal fuel instead of N-Heptan and feared some deviations based on the double converting, but that seems to be fine though :)Quote I've happily been using the 550cc Tvub settings which looking at your pulsewidth table is a little different, so I will try that instead to see if its any better. Nice work though i hear ya, as thats the way i've done it till now too. Maybe worth to mention: look at my axis data. its not the usual 8 and 16v borders, but 9 and 15. But besides that it's quite good to see that the values seem to be shifted a little bit to the right. e.g.: 10V: 550s: 1.43 ; 600s: 1.93 12V: 550s: 1.04 ; 600s: 1.34 14V: 550s: 0.78 ; 600s: 1.0 Maybe a bigger Piston/needle? Edit: If someones interested: made a short clip of one of them squirt into an open cup to see the spray. 10ms pw and 10ms pause here: https://www.youtube.com/watch?v=cHeLIpMXiRk&feature=youtu.be Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: aef on November 19, 2014, 03:56:34 AM Great!
not sure if i understand this correct: row number 3 with 10ms pulsewidth is the one to use 1:1 in TVUB ? Title: Re: Sending a 60lb ev14 out for testing. Would like to send more, but... Post by: ddillenger on November 19, 2014, 08:10:53 AM I just saw this. It is good!
|