I've had a PurolatorOne PL14615 filter on my 2015 Subaru WRX for the past 6,600 km. I decided to change it early. Ever since installing it, it seemed that it took longer for the oil pressure light to turn off during cold starts compared to the previous OG Fram Ultra. Also, on chilly mornings (<10°C), the oil pressure light would sometimes flicker back on for a brief moment after it had turned off.
Before removing the Purolator, I decided to measure how long the oil light took to shut off after a cold start, so that I could compare it to its replacement, a Pentius PLXL7317. I measured the time between first movement of the tachometer to when the oil pressure light shut off. Here are the results:
PurolatorOne PL14615: 0.88 seconds, 20.5 engine revolutions (11°C oil, >24 hours since last engine start)
Pentius PLXL7317: 0.32 seconds, 4.9 engine revolutions (7°C oil, >24 hours since last start)
Pentius PLXL7317: 0.28 seconds, 3.9 engine revolutions (22°C oil, 6 hours since last start)
Pentius PLXL7317: 0.20 seconds, 2.3 engine revolutions (40°C oil, 0.5 hours since last start, the first start after installing the filter, did not pre-fill the filter)
OG Fram Ultra XG3593A: From memory, the light would shut off just as the engine hit 1700-1800 cold idle rpm, which would equate to around 10 engine revolutions.
I expected that the Pentius would do better, since it has around twice as much media, but I didn't expect such a noticable improvement. I suspected the Purolator was too restrictive, and that maybe its spiral spring bypass design didn't perform that well either, so I cut it open to do some forensics. It didn't have any obvious defects, and the filter media was still really clean.
Bypass Valve:
I measured how much the bypass valve opened at various pressures by putting various weights on the centre of the spring. I recorded this on video so I could measure the valve movement from video stills afterwards. There was a very slight opening at 30 psi. I'd estimate the cracking pressure to be between 28 and 30 psi. Purolator rates it at 20-30 psi.
I did some pressure drop calculations based on the size of the opening to estimate the flow rate of oil through the valve at these pressures (with warm oil). On this engine, if the filter was clogged, it seems that the bypass valve dP would reach around 38 psi, and the oil flow rate at high rpm would drop from the nominal 16 GPM that this oil pump can flow, to around 11-12 GPM.
A pressure rise to 10 psi above the cracking pressure does not seem bad, and this bypass valve performs fine by that metric, but the cracking pressure itself is a bit too high for my liking. A lower bypass setting would be a bit safer for the engine if the filter is allowed to clog.
Filter Media:
According to the Purolator datasheets, this filter should have 593 cm^2 of media area, which isn't a whole lot. I measured the useful media area to be only 520 cm^2. The OEM filter for this engine has 900 cm^2, while I'd estimate the Pentius XL 7317 at 1140 cm^2.
In terms of cold flow dP, I'd expect the PL14615 to perform similarly to the Bosch 3330 that Brand Ranks tested. That filter used a similar type of media and had the same media area (but with a much lower bypass setting). It dropped 11.3 psi at 3 GPM. When the oil is this thick, the filter dP seems to increase with the square of the flow rate until the bypass valve opens, so it would hit around 45 psi at 6 GPM if the bypass didn't open (which is this engine's flow rate at cold idle rpm if it's not in pressure relief). With the bypass relieving some pressure, the maximum media dP should end up closer to 35 psi.
So, it seems that this filter could easily exceed the bypass pressure on cold starts, even with the high bypass setting. I'd estimate that the media dP of the Pentius filter would be <10 psi at 6 GPM in a cold flow test, which is under its 13 psi bypass pressure rating. The Purolator's dP would be around 25 psi higher.
I don't really understand why a ~25 psi difference in filter dP would cause such a big change in the time it takes the engine to build oil pressure downstream of the filter. The added restriction would cause the oil pump to relieve more flow, but this engine's oil pump has a relief pressure of 102 psi. 25 psi is a small enough fraction of that, that I wouldn't expect a huge flow reduction.
Filter Canister Restriction:
I measured the inlet holes on the PL14615, and did my best to measure the tiny louver holes in the centre core as well. I did some pressure drop calculations to estimate the dP at a flow rate of 60 L/min with warm oil.
The baseplate dP is 6 psi, and the centre core dP is 3.4 psi, for a total canister dP of 9.4 psi (this doesn't include the filter media). This is more restriction than any other filter I've done these dP estimates for. For comparison, total canister dP would be around 3.6 psi for the Pentius XL 7317, and 2.4 psi for the OEM Tokyo Roki.
The pressure drop across the canister wouldn't be as high in a cold start scenario where flow rates are lower, so I don't think the restrictive canister contributes very much to the Purolator's apparent poor cold flow performance, but it all adds up.
Conclusions
• Don't put a tiny oil filter on an engine with high oil flow, even if the data sheet suggests it has reasonably low dP with warm oil at a low flow rate.
• The Purolator bypass design with the spiral spring seems like it should work fine, despite not opening very far.
• The restriction of these louvered centre cores can start to become significant at high enough flow rates, especially on small filters.
• Pentius XL filters seem to be well designed. They're the only filters I'm aware of that have more media area than the OEM Tokyo Roki filters, and should work well on Subarus despite the lower bypass setting.
I might post some C&P photos when I have time.
Before removing the Purolator, I decided to measure how long the oil light took to shut off after a cold start, so that I could compare it to its replacement, a Pentius PLXL7317. I measured the time between first movement of the tachometer to when the oil pressure light shut off. Here are the results:
PurolatorOne PL14615: 0.88 seconds, 20.5 engine revolutions (11°C oil, >24 hours since last engine start)
Pentius PLXL7317: 0.32 seconds, 4.9 engine revolutions (7°C oil, >24 hours since last start)
Pentius PLXL7317: 0.28 seconds, 3.9 engine revolutions (22°C oil, 6 hours since last start)
Pentius PLXL7317: 0.20 seconds, 2.3 engine revolutions (40°C oil, 0.5 hours since last start, the first start after installing the filter, did not pre-fill the filter)
OG Fram Ultra XG3593A: From memory, the light would shut off just as the engine hit 1700-1800 cold idle rpm, which would equate to around 10 engine revolutions.
I expected that the Pentius would do better, since it has around twice as much media, but I didn't expect such a noticable improvement. I suspected the Purolator was too restrictive, and that maybe its spiral spring bypass design didn't perform that well either, so I cut it open to do some forensics. It didn't have any obvious defects, and the filter media was still really clean.
Bypass Valve:
I measured how much the bypass valve opened at various pressures by putting various weights on the centre of the spring. I recorded this on video so I could measure the valve movement from video stills afterwards. There was a very slight opening at 30 psi. I'd estimate the cracking pressure to be between 28 and 30 psi. Purolator rates it at 20-30 psi.
I did some pressure drop calculations based on the size of the opening to estimate the flow rate of oil through the valve at these pressures (with warm oil). On this engine, if the filter was clogged, it seems that the bypass valve dP would reach around 38 psi, and the oil flow rate at high rpm would drop from the nominal 16 GPM that this oil pump can flow, to around 11-12 GPM.
A pressure rise to 10 psi above the cracking pressure does not seem bad, and this bypass valve performs fine by that metric, but the cracking pressure itself is a bit too high for my liking. A lower bypass setting would be a bit safer for the engine if the filter is allowed to clog.
Filter Media:
According to the Purolator datasheets, this filter should have 593 cm^2 of media area, which isn't a whole lot. I measured the useful media area to be only 520 cm^2. The OEM filter for this engine has 900 cm^2, while I'd estimate the Pentius XL 7317 at 1140 cm^2.
In terms of cold flow dP, I'd expect the PL14615 to perform similarly to the Bosch 3330 that Brand Ranks tested. That filter used a similar type of media and had the same media area (but with a much lower bypass setting). It dropped 11.3 psi at 3 GPM. When the oil is this thick, the filter dP seems to increase with the square of the flow rate until the bypass valve opens, so it would hit around 45 psi at 6 GPM if the bypass didn't open (which is this engine's flow rate at cold idle rpm if it's not in pressure relief). With the bypass relieving some pressure, the maximum media dP should end up closer to 35 psi.
So, it seems that this filter could easily exceed the bypass pressure on cold starts, even with the high bypass setting. I'd estimate that the media dP of the Pentius filter would be <10 psi at 6 GPM in a cold flow test, which is under its 13 psi bypass pressure rating. The Purolator's dP would be around 25 psi higher.
I don't really understand why a ~25 psi difference in filter dP would cause such a big change in the time it takes the engine to build oil pressure downstream of the filter. The added restriction would cause the oil pump to relieve more flow, but this engine's oil pump has a relief pressure of 102 psi. 25 psi is a small enough fraction of that, that I wouldn't expect a huge flow reduction.
Filter Canister Restriction:
I measured the inlet holes on the PL14615, and did my best to measure the tiny louver holes in the centre core as well. I did some pressure drop calculations to estimate the dP at a flow rate of 60 L/min with warm oil.
The baseplate dP is 6 psi, and the centre core dP is 3.4 psi, for a total canister dP of 9.4 psi (this doesn't include the filter media). This is more restriction than any other filter I've done these dP estimates for. For comparison, total canister dP would be around 3.6 psi for the Pentius XL 7317, and 2.4 psi for the OEM Tokyo Roki.
The pressure drop across the canister wouldn't be as high in a cold start scenario where flow rates are lower, so I don't think the restrictive canister contributes very much to the Purolator's apparent poor cold flow performance, but it all adds up.
Conclusions
• Don't put a tiny oil filter on an engine with high oil flow, even if the data sheet suggests it has reasonably low dP with warm oil at a low flow rate.
• The Purolator bypass design with the spiral spring seems like it should work fine, despite not opening very far.
• The restriction of these louvered centre cores can start to become significant at high enough flow rates, especially on small filters.
• Pentius XL filters seem to be well designed. They're the only filters I'm aware of that have more media area than the OEM Tokyo Roki filters, and should work well on Subarus despite the lower bypass setting.
I might post some C&P photos when I have time.
