Sorry I have nothing constructive to add, but geesh bigtinboat I finally see somebodys fingers dry cracked more than mine... ouch
Dave
Dave
Upvote 0
FX Crown: The Dreaded "Regulator Creep"
- Thread starter MP44
- Start date
Hi Rob, thanks for the extra info and photo of the washer stack. I have not rebuilt one of them so that gives me a better idea, and I'm inclined to think you, Chuck, and perhaps most significantly, Robert Lane, are correct about the sloppy Belleville fit being a problem. Certainly seems very likely to lead to variation in the setpoint, although it's not immediately obvious to me why it causes creeping unless it's so bad that the piston is tilting slightly. It wouldn't have to tilt much since we are dealing with a fairly rigid seal material.
Upvote 0
Hi Weatherby, thanks for the link to your regulator mod tutorial. I think it's great that you've found a solution to the problem, and I hope I don't come across as pedantic about this but I do not understand how this mod serves to slam the door harder.
I will attempt to step through the logic so perhaps you can point out to me where I'm going astray. So let's compare the behavior of stack 1 and stack 2:
stack 1: low spring rate, long travel
stack 2: high spring rate, short travel
From what I gather reading through these threads, the target setpoint is about 150 bar in either case. For a given setpoint, either arrangement will have the same force of X pounds compressing the washer stack...which must be true for them to both reach 150 bar.
The difference is that stack 1 will have compressed a greater distance to get there.
What we are keenly interested in is how the two stacks behave just as the flow of air is about to be halted. At this threshold where air is slowly creeping past the seat, pressure must continue to build a bit more until it sufficiently pushes the spool against the seat to completely cut off the flow of air. Whatever that distance is, stack 2 will need more of a pressure change to move that distance. Conversely, stack 1 will need a smaller pressure change to move that distance...which makes this arrangement quicker to slam the door.
Is it perhaps instead that 150 bar is at the upper limit, that the washers are being squeezed almost completely flat and are becoming nonlinear?
I will attempt to step through the logic so perhaps you can point out to me where I'm going astray. So let's compare the behavior of stack 1 and stack 2:
stack 1: low spring rate, long travel
stack 2: high spring rate, short travel
From what I gather reading through these threads, the target setpoint is about 150 bar in either case. For a given setpoint, either arrangement will have the same force of X pounds compressing the washer stack...which must be true for them to both reach 150 bar.
The difference is that stack 1 will have compressed a greater distance to get there.
What we are keenly interested in is how the two stacks behave just as the flow of air is about to be halted. At this threshold where air is slowly creeping past the seat, pressure must continue to build a bit more until it sufficiently pushes the spool against the seat to completely cut off the flow of air. Whatever that distance is, stack 2 will need more of a pressure change to move that distance. Conversely, stack 1 will need a smaller pressure change to move that distance...which makes this arrangement quicker to slam the door.
Is it perhaps instead that 150 bar is at the upper limit, that the washers are being squeezed almost completely flat and are becoming nonlinear?
Upvote 0
Interesting discussion
Ofcourse it is all very theoretically, and trying to understand what is going on inside a regulator regarding materials and physics is hard to explain.
My theory (doesn't have to be I'm right ofcourse) is that the Delrin material of the regulator piston contributes to the total spring stack of the whole system, and the flexibility of the Delrin plays a important role in this process.
Delrin is relatively soft (but pretty hard for a "plastic"), thus can be compressed because of it's flexability. The BV washers have fairly sharp edges, which could dig into the soft Delrin, and the Delrin is starting to act like a spring as well.
Because of this reason, I added a regular washer to divide the springforce generated by the spring stack over a bigger surface.
At the sealing end, this flexibility also seems to play a role, because the regulator adjustment screw also slightly "digs into" the sealing surface of the Delrin regulator piston to make it seal.
With a higher spring rate (f.i. with double stacked BV washers), the spring force increases about twice as fast for the same travel/movement.
And in my opinion, that is where this phenomenon comes into play. Because the force/travel ratio is higher, the "sealing barrier" of the Delrin is getting less important and hence also less prone to creeping. (Less piston travel for the same spring force)
Man, it is hard to put my thoughts on paper, I hope I am making myself a bit understandable
Cheers,
Gijs
Edit: You are right that in stock form of older gen. Crowns, the spring stack is compressed all the way. Result is that the BV washers are starting to act like regular washers, causing the reg. to overshoot. Newer Crowns have bigger and thicker BV washers, 10x4.2x0.6mm compared to 8x4.2x0.4mm for first gen. Crowns. The housing is changed physically as well to make room for the bigger BV washers.
Upvote 0
Weatherby, thanks for taking the time to address my questions. You put forth two aspect of the mod so I’d like to comment on them in the same order.
The first part is really quick because we agree ?. I think I follow you on the part about using a flat washer to keep the Bellevilles from digging into the regulator body. No doubt in my mind that’s a good idea. At best, wear in that area would cause the setpoint to drift over time. At worst, the wear would be uneven and potentially contribute to the piston tilting slightly and lead to the creeping problem.
However I still disagree with the reason why doubling the spring rate is helping things. I’m not disagreeing that it helps, just disagreeing with the reason. There is a minimum amount of interference that must occur at the sealing end—the degree to which the Delrin seal and valve seat must squish together, if you will—before the flow of air becomes halted.
This amount of “squish” necessary for this to occur is dictated by how well these two surfaces mate together. It does not know or care about the washers out there that are pushing back and providing a slowed approach. It cares only that there is some pressure at which this amount of squish is achieved…and that pressure is 150bar.
There is a difference, however, in the way the stack #1 and stack #2 provide for the slowed approach. Stack #1 needs less of a pressure rise to move X distance, thus it should coast on in and make the seal more easily. By this reasoning, I would expect stack #2 to be more apt to creep. In other words, just the opposite of your theory.
That’s why I was brainstorming other reasons for the improvement, like the possibility that the single Bellevilles are becoming nearly flattened at 150bar and behaving poorly as a result. Or perhaps that the slop between the Bellevilles and the spool is encouraging the spool to tilt ever so slightly, preventing it from mating neatly to the valve seat the way it should. High pressure air is very skilled at moving through the tiniest of gaps so it wouldn’t take much.
Lastly, I just want to comment on the desirability for the spring stack to have as much travel as is reasonably possible. It helps both the repeatability and the recovery time. That is why a simple series arrangement () () () () () is so commonly used. The spring rate of this stack of 10 is the same as the that of a single washer. Meaning theoretically we need only a single washer to achieve a particular setpoint. But the dynamic operation of the assembly would be so sensitive to minor variations in tolerances and temperature that it would not behave in a repeatable fashion. Oh and the cycle time would be painfully slow.
The first part is really quick because we agree ?. I think I follow you on the part about using a flat washer to keep the Bellevilles from digging into the regulator body. No doubt in my mind that’s a good idea. At best, wear in that area would cause the setpoint to drift over time. At worst, the wear would be uneven and potentially contribute to the piston tilting slightly and lead to the creeping problem.
However I still disagree with the reason why doubling the spring rate is helping things. I’m not disagreeing that it helps, just disagreeing with the reason. There is a minimum amount of interference that must occur at the sealing end—the degree to which the Delrin seal and valve seat must squish together, if you will—before the flow of air becomes halted.
This amount of “squish” necessary for this to occur is dictated by how well these two surfaces mate together. It does not know or care about the washers out there that are pushing back and providing a slowed approach. It cares only that there is some pressure at which this amount of squish is achieved…and that pressure is 150bar.
There is a difference, however, in the way the stack #1 and stack #2 provide for the slowed approach. Stack #1 needs less of a pressure rise to move X distance, thus it should coast on in and make the seal more easily. By this reasoning, I would expect stack #2 to be more apt to creep. In other words, just the opposite of your theory.
That’s why I was brainstorming other reasons for the improvement, like the possibility that the single Bellevilles are becoming nearly flattened at 150bar and behaving poorly as a result. Or perhaps that the slop between the Bellevilles and the spool is encouraging the spool to tilt ever so slightly, preventing it from mating neatly to the valve seat the way it should. High pressure air is very skilled at moving through the tiniest of gaps so it wouldn’t take much.
Lastly, I just want to comment on the desirability for the spring stack to have as much travel as is reasonably possible. It helps both the repeatability and the recovery time. That is why a simple series arrangement () () () () () is so commonly used. The spring rate of this stack of 10 is the same as the that of a single washer. Meaning theoretically we need only a single washer to achieve a particular setpoint. But the dynamic operation of the assembly would be so sensitive to minor variations in tolerances and temperature that it would not behave in a repeatable fashion. Oh and the cycle time would be painfully slow.
Upvote 0
I have and FX crown and mine is doing the same creep motion. Wayne Linsenbach 6 old farm road Hummelstown pa 17036. Thankyou
Upvote 0
I would not even worry about some creep over a peroid of time.. The higher you go, the more creep.. It is physics in the design of a Regulator that has a external ajusting screw that seals the under air pressure as it rises..
Upvote 0