Baldwin bypass and a challenge

[SWHeat]

Follow the multiple links from the threads in the sticky at the top of this bypass forum. Indeed, Blackstone's only real number if I remember correctly is the 10 micron number. The rest of the numbers are derived from a mathematical formula.

Again, follow the links in the sticky. Who knows some links may not work anymore due to the software upgrade. I have not bothered to look lately.

 

[Gary Allan]

Originally Posted By: gto78
Originally Posted By: Gary Allan
I emailed Ryan at B-S a while back about this. This is a machine that calculates the particle count by a measured amount of oil pushed through a membrane. The upramp in pressure as well as the rate of upramp then generates a particle count based on contouring models that were programmed using the optical method.

So, it is a "facsimile" of a particle count. It's what you do with opaque fluids. The other method, IIRC, is to use a dilution factor and then apply the proper multiplier.


I didn't really understand what you just said....lol Are one of those methods giving results by using a math formula instead of "real" particle counts? What UOA lab is better in your opinion? I'd like to start sending in UOA's to see how I compare, but I'm not wasting my money and sending it to a less than accurate lab.


You can't get an optical read on opaque fluids. The B-S method should NOT be a calculated process (by a person). Just email Ryan and ask him for the machine man's name and google it. I did this some time ago...but don't have the info handy.

 

[SWHeat]

Originally Posted By: Gary Allan
You can't get an optical read on opaque fluids. The B-S method should NOT be a calculated process (by a person). Just email Ryan and ask him for the machine man's name and google it. I did this some time ago...but don't have the info handy.


If the screen is set at 10 microns, then just how do they get the other numbers? Math my friend, math. They may tell you from many years of data collection this is the distribution curve that is normal with a given sum at 10 microns. Still it is not an actual data set at 2 micron, 5 micron, 10 micron, 15 micron, etc. Not happening. Does it really matter if the mathematics are done by a person or a machine?


And yes, you cannot get an optical scan on opaque fluids. Diesel oil with soot is too black to get a reading with an optical scanner. Many gas engine oils are capable of being optically scanned, especially at low mileages (5K or less) with a non-sludged engine. Terry Dyson can do it for an extra fee.

 

[Gary Allan]

Quote:
Not happening. Does it really matter if the mathematics are done by a person or a machine?


I don't think you understood the calculation method of the machine as it was described to me. I saw no authoritative source for the 10um membrane.

Surely anyone who's dealt with PID loops can see how varied Proportion - Interval = and Duration will result in a given curve of input in a process control situation. Now it's surely not simple "math" to reintegrate a curve into a given model of particle distribution.

That's the process as I undertand it. I could attempt to articulate how one might derive such a varied result from modeling ..but it may have nothing in common with the actual way they reason it.

..but if it's junk ..then more than a few labs are spending LARGE for nothing. Those machines aren't cheap.

 

[SWHeat]

Originally Posted By: Gary Allan
Quote:
Not happening. Does it really matter if the mathematics are done by a person or a machine?


I don't think you understood the calculation method of the machine as it was described to me. I saw no authoritative source for the 10um membrane.


My memory may be incorrect on the exact size of the membrane/screen. The results are the same. One data point. I need not understand the complexity of the math to know that it is all being derived from one data point.

Quote:
Surely anyone who's dealt with PID loops can see how varied Proportion - Interval = and Duration will result in a given curve of input in a process control situation. Now it's surely not simple "math" to reintegrate a curve into a given model of particle distribution.

That's the process as I undertand it. I could attempt to articulate how one might derive such a varied result from modeling ..but it may have nothing in common with the actual way they reason it.

..but if it's junk ..then more than a few labs are spending LARGE for nothing. Those machines aren't cheap.


Hey, when you have pitch black diesel oil, you have no other choice. Some data is better than none. You've heard it before....one man's junk is another man's treasure...

 

[Gary Allan]

I would assume that the manufacturers of the pore blockage machines have already done this, but I imagine that if one has two identical samples of transparent composition and subject one to the preferred optical method and one to the pore blockage method, one would see how good or poor the modeling is.

I don't have a dog in the fight, if you see what I'm saying. I just would find it hard to sell someone a machine that is as useful as udders on a bull. I would hope that most labs would have a hard time buying one too.