mx5cartalk.com

Following on from Madjack's comment about oil pressure drop under hard braking I thought I would have a look at my recent data.

First off a complaint about the Racepak software (Datalink II) it is nowhere near as good as Innovate's Logworks. When it exports the export is based on the format being displayed, NOT the raw data. For example, if G's are displayed to 1 decimal place that is how it will be exported not 3 decimal places.... Therefore it is REALLY painful to have to export then make the graphs in Excel.

But, here is oil pressure and lateral G's. Oil pressure does not drop below 30 PSI irrespective of G's. However as the Sprint layout is mainly right handers this is where the majority of the data points are.

QR_Press by Eipeip, on Flickr

When oil temp and pressure are graphed by session it shows that oil pressure is trending down session after session (heat soak). This also corresponds with the oil pressure dropping to its lowest. Based on this data I would say that the low pressure is a result of heat soak and not G's (at least at my G levels). Notice oil temp never get up to even 95°C! The PWR radiator is very effective

QR_Oil_Temp by Eipeip, on Flickr

Again a tyre plot shows that I'm not trail braking very much rather either braking or turning not both. Session 1 I was but stopped doing it. Need to look some more.
QR_Tyre by Eipeip, on Flickr

Thinking about Gladiators spark plug failure I mapped EGT's/RPM. At the most there is about 40°C between EGT's so happy with this as well.

QR_EGT by Eipeip, on Flickr

My oil pressure wasn't dropping less than 25 PSI but it was a definite dip immediately after hard braking from my usual 50-60PSI at temp. Due to the Peterson valve we both run, the oil pressure usually sits fairly static at whatever the Peterson valve is set to regardless of the revs. The pressure will drop a little as the oil temperature makes the oil thinner which is what you are seeing.

My concern wasn't actually related to the dip to 25/30PSI... an engine will run fine at less than that. My concern was that the dip was being caused by the sump pickup sucking air which means it was drawing in air bubbles. That air then travels through the oil cooler and is fed back into the engine probably right when the engine is back under full load. That was the reason I didn't think an accusump would help because it would just fix the pressure without preventing the air.

I'll download some logs and show you my before and after the sump baffle so you can see how bad the dips were.

Thanks Madjak before and after data would be appreciated

A little off topic, but managed to merge the Vbox video with the Datalink II information. Needs a bit of tidy up, nice way to spend a Sunday making new input templates for Dashware and learning what Datalink II cannot do!

That's some cool data magpie!

You can make all those graphs in i2 pretty easily so you don't need to export anything.

Sent from my SM-N9005 using Tapatalk

Thanks Shane, the Datalink II software basically sucks. Logworks was so much better as is the Aim Software.

The g plot shows that I have a lot more track time to do. Good article http://www.temporal.com.au/ggdiag.htm

Went back and looked at the data from Morgan Park (SDT Day).

There were a few low oil pressure dips, and at first I was very worried.

Session 1 (red dots) on the positive side were recorded when I did a spin. Further the RPM was around 1,200 for the duration of these 'dots'. When RPM increased so did the oil pressure.

Session 4 (purple dots) these were much more alarming as it did not involve a spin. Rather, for some reason I decided to ease off and coast down to T4 doing a cool down lap, however when lifting off into T4 the RPM dropped to below 1,000 and as a consequence so did oil pressure.

Therefore the events of low oil pressure were at very low RPM and when RPM increased so did the oil pressure. May still look at Madjak's baffle just to be safe.

MP_Oil_Press by Eipeip, on Flickr

Pink - Oil Pressure, Blue - RPM, Red - Speed
MP_T4 by Eipeip, on Flickr

Oil pressure/temp shows the same trend of pressure dropping as heat soak increases.
MP_Oil_Temp by Eipeip, on Flickr

G Plot shows a little better on right hand corners, than on the Sprint, but again not working to the limits of grip.
MP_G_Plot by Eipeip, on Flickr

The dips you're getting in the pink line are similar to mine, maybe just a little less severe... I'm talking about the ones at 57 seconds, 67 seconds, 75 and 80 where the oil drops low then back up again whilst your revs are higher than 4000. If you look at the start of the log, the engine should be making 60PSI @ 4000 rpm, but the dips above all occur during heavy braking zones and the oil pressure is basically halving.

To reproduce the issue all you need is sticky tyres (ie R-comps or better), high revs beforehand (so that oil accumulates in the head) and hard braking. I think any BP engine will see the same issue if you do all three especially if running any aero.

Whether or not it's an issue depends on how you drive I think. If you like blipping the throttle under brakes (like me) then you could easily hit over 8000 rpm during one of these low pressure dips you'll could be running on the metal of the bearing. Do it enough and that bearing could fail. Given it's so easy to add the baffle plate like mine I wouldn't hesitate to install it during a rebuild if you see these dips. I don't see any negatives to running the baffle plate. It's not a critical issue, but I feel much better with it resolved... I know if I do a bearing now that there are other issues.

Thanks Madjak, agree that drivers have a role to play and I'm not pushing mine like you are. Those dips you noted are at gear changes and more than likely under heavy braking as well. Irrespective it may be worth looking into your baffle when the engine is out for an inspection.

Played some more with the data. Plot with RPM and Oil pressure. Again this shows that the low oil is only happening at low RPM. A limitation with excel is 32,000 data points and the graph was trying to use 69,000.

MP_Oil_RPM by Eipeip, on Flickr

Histogram of oil pressures. The log rate is 10mHz so the below 20 PSI account for 2.4 secs in 6,942.2 secs of data.

MP_Hist by Eipeip, on Flickr