Another idea for a drop-in replacement for the PRV
|
I have been thinking for a long time about a drop in replacement PRV design that actually works to regulate pressure rather than relieve pressure and operates smoothly through a range of RPM’s rather than just snapping open and closed.
Here is my idea, and before I get going, I am really hoping that either BobP, and/or David Amsler, since they have rigs they built to test such things, and have some machining skills, they may seriously consider building a prototype of my design to let me know if it works.
The basic concept:
Jay
|
|
|
|
Mick O'Connor
Jay, I’ve been testing a PRV for Bob Panther. Works great and regulates OP extremely smoothly. Mick
On 7 Feb 2021, at 21:30, Jay Scheevel <jay@...> wrote:
|
|
|
|
David Amsler
That is basically a sleeve valve, but the small inlet shown retains the problem with the Jabiru design, Only a small area of the end of the piston is exposed to manifold pressure, but when it opens, the larger face is exposed causing the piston to be rapidly pushed back further. The tapered nose and seat can be eliminated, with a piston made the same dia. as the original exposed nose and prevented from moving too far forward (and falling out) by cross pin at inlet or at tail end of piston. I also think you will find it impossible to find a spring that small dia. that has a low enough spring rate for this. My design actually uses the original Jabiru spring, Bob's is a custom spring. Bob's is installed and building hours, I have had too many diversions in the last year. Hope to install it when the weather improves. Experimented with stepped progressive opening of the side holes and found that the quicker all open, the less prone to pulsation. Not sure just why that is. Bob's design and mine differ slightly in design of the side holes, but both seem to work. His is easier to machine. Your exit ports should total a bit more area than your inlet. Also need to take into account the pulsations from the pump. If given some pre-travel, the piston will be moving without opening as pressure builds, and without pre-travel it will be opening too soon resulting in poor pressure regulation. Fortunately, given the movement the pump causes, the PRV is the second best lubricated part in the engine. I have also made a PRV that incorporates a hydraulic damping feature, and it worked great in the test rig, but looks to be too many places for Murphy's law to operate where a particle in the oil could hang it up, so abandoned that approach. David A.
On Sun, Feb 7, 2021 at 4:30 PM Jay Scheevel <jay@...> wrote:
|
|
|
|
Hi David,
Thanks for the comments. You will note that the piston does not relieve any pressure until well after the entire face of the piston is exposed. When if first opens, it is still occluding the first relief port, so no pressure effect. It may take a very small pressure to push the piston back from the seat initially, but that does not reduce the oil pressure until it opens up a port, which is after some sliding.
Jay
From: main@JabCamit.groups.io <main@JabCamit.groups.io> On Behalf Of David Amsler
That is basically a sleeve valve, but the small inlet shown retains the problem with the Jabiru design, Only a small area of the end of the piston is exposed to manifold pressure, but when it opens, the larger face is exposed causing the piston to be rapidly pushed back further. The tapered nose and seat can be eliminated, with a piston made the same dia. as the original exposed nose and prevented from moving too far forward (and falling out) by cross pin at inlet or at tail end of piston. I also think you will find it impossible to find a spring that small dia. that has a low enough spring rate for this. My design actually uses the original Jabiru spring, Bob's is a custom spring. Bob's is installed and building hours, I have had too many diversions in the last year. Hope to install it when the weather improves.
Experimented with stepped progressive opening of the side holes and found that the quicker all open, the less prone to pulsation. Not sure just why that is. Bob's design and mine differ slightly in design of the side holes, but both seem to work. His is easier to machine. Your exit ports should total a bit more area than your inlet. Also need to take into account the pulsations from the pump. If given some pre-travel, the piston will be moving without opening as pressure builds, and without pre-travel it will be opening too soon resulting in poor pressure regulation. Fortunately, given the movement the pump causes, the PRV is the second best lubricated part in the engine.
I have also made a PRV that incorporates a hydraulic damping feature, and it worked great in the test rig, but looks to be too many places for Murphy's law to operate where a particle in the oil could hang it up, so abandoned that approach.
David A.
On Sun, Feb 7, 2021 at 4:30 PM Jay Scheevel <jay@...> wrote:
|
|
|
|
jabcamit@...
Hi Jay,
I'm happy that my design does the job - regulates rather than oscillates, is reliable, and is easy (cheap) to make. Many thanks to David for bouncing ideas back and forth and searching for any design weak points, and to Mick for testing this design, and a few previous ideas, in a real engine - measurements of pressure and pressure ripple over ranges of rpms at different oil temperatures, life testing etc. Ignoring valve flutter (Bernoulii effect?) the key to stability is that at all stages of valve opening there is an overall negative feedback. That is the more the valve opens the more resistance (via the spring) there is to its movement. If this is not the case then damping will be required to keep the valve stable. But as David says this increases complexity, cost, and perhaps risk of failure. A positive feedback component is introduced if the area of all moving parts that face the oil flow (the exposed area of the plunger and moving bits of the spring ) increases at any stage of the valves opening. Positive feedback is also caused by the viscous drag of the oil moving past moving parts. A second consideration is to ensure sufficient oil can flow through the PRV when it is open. I did static oil flow tests on my design to ensure that it offered less resistance to oil flow than the Jabiru PRV. Worse case is the engine revving when the oil is really thick and cold (eg if the throttle is blipped before the oil warms up). BobP
|
|
|
|
David Amsler
Bob, Minor point, but as you know, our sleeve valve PRVs do move with the pump pulses. The hydraulic damping did eliminate that as well as any tendency to hammer or buzz, but so long as the valve design does not let the valve to bottom out and start "buzzing" from the pump pulses, that movement looks to be harmless in the well lubricated environment so the added complexity and increased modes for failure make it not worth the risk for that action either. Also the design of our sleeve valves the movement that does occur is in the sleeve, which is a separate part rather than moving within the bore of the engine block, so it is easy to inspect and if ever necessary, to replace. David A.
On Thu, Feb 11, 2021 at 12:46 PM <jabcamit@...> wrote: Hi Jay,
|
|
|
|
jabcamit@...
Hi David & thanks for mentioning those points (buzz and plunger moving with the pump pulsing).
BUZZING My PRV didn't buzz on my test rig but it might be different in a real working engine. Its not possible to test (ie listen for) for any buzzing in Mick's running engine, but if it does buzz it has no discernable effect on pressure regulation, pressure pulsing, or PRV part wear (at 50 hours). PLUNGER FOLLOWS PUMP MODULATION By design my PRV uses low inerta moving parts so I'm sure it will follow the pressure modulation generated by the pump. The plunger movement should tend to reduce that modulation (valve opens more as the peaks are approached and the increased oil PRV flow decreases those peaks). I can't quantify this improvement if any. If the inertia/ resonant frequency of the PRV gets 180 degrees out of phase with the pressure pulses it could make things worse. It would be easy to quantify the improvement (or otherwise) by running a representitive test rig (I think you have one) with my PRV design and then a fixed aperture bipass in the PRV hole; and then comparing the modulation (acmV) for the two setups. WEAR As you speculated, wear was minimal after 50 hours operation in Mick's engine - some polishing between bore and plunger but no measurable wear. Based on the 50 hour wear I think my PRV should remain servicable for at least 1000 hours. But in any event deterioration would be gradual - oil pressure (hot engine, low rpm) would just slowly decline over the years. No catastphic failure (not like a popped oil cooler due to Jabiru PRV hammering), no wear in the engine PRV cavity, just pop in a replacemnt PRV. Compared with the other plunger/bore mechanisms in the engine (pistons in cylinders) the PRV plunger movement is minute just a few mm, side loads are minimal, temperature fluctuations are minimal, and lubrication is several orders better. I think new pistons and cylinder may be required long before the PRV wears out. BobP
|
|
|
|
David Amsler
Bob, I think we traced the "buzzing" to pre-travel, and when mine was modified to about the same pre-travel as yours I could not make it buzz. I tried that with both heavy and lightweight pistons, but would have to go back in my records to see if that made any difference, off hand do not recall that it did, so think we have gone as far as we can with the test rig. Real engines may prove us wrong, but at this time, the sleeve valves we made seem to be the way to go. At the moment, the test rig is dismantled and the pump is back on old engine under a pile of stuff in the un-insulated steel barn which I would have to dig a path through the snow to even enter. (Not in any rush to do so.) David A.
On Fri, Feb 12, 2021 at 7:11 AM <jabcamit@...> wrote: Hi David & thanks for mentioning those points (buzz and plunger moving with the pump pulsing).
|
|
|