Page 5 of 8

Re: digital steam engine indicator

Posted: Fri Oct 07, 2016 1:46 am
by barts
Our traditional cards are right; it's using crank rotation w/o correction instead of linear piston position that's wrong.

A small rack moved by the cross head and high speed encoder would work: ... Auy18P8HAQ

You could use a strong magnet to hold the encode & rack (w/ ball bearings guiding it) to the base of the engine along side the crosshead; you could drive it from the crosshead w/ a piece of stiff music wire.

For a permanent installation, a linear encoder w/ sufficient speed may be available, although most
seem to max out at 40-50 inches/sec; w/ a 1/2" diameter pinion + rack the encoder above could handle
more than 3 x times that.

- Bart

Re: digital steam engine indicator

Posted: Fri Oct 07, 2016 3:13 am
by troye_welch
Looking back a few posts, I don't think much of a home made linear potentiometer. The environment is too grubby for any sort of exposed ohmic contact
I guess you can go your way and I'll go mine then. I change rotary pots out of German-designed multi-million dollar medical diagnostic imaging equipment quite regularly. In fact, the 1K and 2K 15-turn 1-watt pots are the most common point of failure in these, and they get way fewer cycles than you would at an hour at 200 RPM. I never suggested a permanent installation, so your "grubby environment" comment is moot.

Furthermore, why would you want to have a continuous indicator running? Just a cool sort of gauge next to the tach and pressure gauge? A digital display like that would look pretty out of place next to analog-everything-else. I already said my main motive was for adjusting valve timing. The old timers would get one diagram at a time to either check/verify proper operation, or as a starting point for trial and error adjustments. You have to shut your engine down anyway to make valve adjustments, so it's not like you can adjust it on the fly. A well-implemented, continuously running indicator would require a lot more processing power and a high-(ish)-res display to boot as well as a power supply and further complications.

A rack/pinion on a multi turn rotary pot or an encoder could work, but the pinion gear would have to be large enough to not spin the measuring device at insane speeds, if it was to last more than a few minutes.

A simple length of nichrome resistance wire (I have a 5 pound spool), a pressure transducer, two 8-bit A/D converters, and a low powered microcontroller is all that's required. I've got all of those on hand. Just have to wire it up and write the code (if my day job left me any spare time). I'll race you to an actual implantation and then when I do it, you can tell me how it won't work. Lol! :lol:

Re: digital steam engine indicator

Posted: Fri Oct 07, 2016 3:58 am
by Lopez Mike
I will be glad to see your slide pot work out. I suspect that by the time you get this all sorted out you may wish that you had just hooked up an antique indicator and cranked out some paper cards.

I don't have any particular need to adjust my valve gear on my boat. But I have lusted after an instrument to tell me what is going on under different loads and speeds. I agree totally that having some high tech mess visible in the engine area would be esthetically wrong but there are ways to hide things.

I have a bicycle speedometer coupled to my engine and calibrated to read r.p.m. The pickup is hard to see down at the flywheel and the speedo is tucked away where it isn't obvious. I have an old brass gauge housing that is slated to hold the speedo/tach.

I would love to arrange a way to measure the torque as part of the permanent installation. Either flexible engine mounts or a strain gauge on the prop shaft. Then, combined with a real time integration of the IHP I would have a readout of engine output and efficiency. Of course not needed but since when are we justifying our hobby on practical grounds???

As Bart can testify, my boat is a poster child for simplicity and reliability. All I need now is one of those caps like Bogart wore. And I liked the arm garters.

Re: digital steam engine indicator

Posted: Fri Oct 07, 2016 12:37 pm
by DetroiTug
High speed ball bearing rotary encoders would work well in this application. US digital in Washington state is a good source for high quality low cost units. For very high speed applications, lower resolution can be selected to help out slower processing speeds. i.e. 25 line encoders. Keep in mind though that 25 lines is actually 25 cycles in quadrature, so that would be 100 output pulses or "edges" per revolution.

Another way is to make your own linear encoder with index. Attached to the crosshead, A strip with holes drilled along it's length and an additional hole or window along side each end window. Two I/R emitters and collectors mounted on a stationary horseshoe frame side by side shooting through the strip. One measuring travel and the second one to sense the end of stroke.

The I/R sensors shooting through a window is common practice on CNC lathes to sense spindle position and speed for threading applications. And they are capable of sensing speeds of 10,000 RPM and more. This is how most all encoders work. Same principle, many different configurations.

An Arduino controller should be able to handle this at low speeds and with short program length.


Re: digital steam engine indicator

Posted: Fri Oct 07, 2016 1:58 pm
by Lopez Mike
The trick will be to cram this all into a 4" brass gauge body so as to fit in esthetically. The pressure and motion sensors could be boxed in with some sheet brass.

The torque sensing might be tricky. I don't know if the various forces on my prop shaft can be isolated from each other so that I would only sense twisting. Probably easier to have a torque arm on the engine with slightly resilient mounts.

Given the way my engine thrashes about at higher speeds there would have to be some serious averaging!

I can see it now. A multiple line LCD readout of rpm, BHP, IHP and efficiency and a skipper who runs aground looking at it.

Re: digital steam engine indicator

Posted: Fri Oct 14, 2016 6:13 am
by fredrosse
"So in our design of a digital indicator we had best stay with sensing piston travel to avoid all the horrors of needing to know the connecting rod length."

The calculation to determine piston position as a function of crank angle is not too difficult, even when considering finite length connecting rods. The axial position (along the main centerline of the engine, passing from the crankshaft center along the axis of piston motion) of the crankpin is the Cosine of the crank angle x the crank radius. The crankpin displacement away from the centerline is the Sine of the crankshaft angle x the crank radius.

For a right triangle (90 degree angle between two sides), A^2 + B^2 = C^2, where A and B are the sides of the triangle adjacent to the 90 degree angle, and C is the hypotenuse (longest side) of the right triangle. The connecting rod length always forms the hypotenuse of a right triangle in this evaluation, and one side of this triangle is always crankpin displacement away from the centerline (the Sine of the crankshaft angle, multiplied by the radius of the crank), call this side "B".

So, one can find the length of "A" (the axial effective length of the conrod between the crankpin and the crosshead) by solving this simple equation: ( C^2 - B^2 ) ^ 0.5 = A. Adding the axial position of the crankpin (the Cosine of the crank angle x the crank radius), plus the axial effective length of the conrod (A), you now have the position of the crosshead pin for any crankshaft angle.

Re: digital steam engine indicator

Posted: Fri Oct 14, 2016 7:11 am
by Lopez Mike
So three native american mothers are sitting around at a powwow talking about their kids. One of them is sitting on part of a hippo hide that her son in the Peace Corps sent her from Africa. The other two are on ordinary cured skins.

One of them brags that her junior high aged son weighs 150 pounds. The second one says that her kid weighs about the same. The mom with the son in the Peace Corps laughs and say that she weighs 300 lbs. As much the two kids together.

Which proves that the sons of the squaws on the other two hides equal the squaw of the hippopotamus hide.

So there. I know geometry too!

Re: digital steam engine indicator

Posted: Mon Nov 21, 2016 9:13 am
by cyberbadger
Would this provide fast enough frequency response for a pressure transducer to capture up to 1000 strokes/min=500rpm up to 200psi? They also sell a "Pressure snubber for steam and gases" for it.



P.S. Would this sensor be able to have meaningful data and to plot for what some are talking about in this thread.

Re: digital steam engine indicator

Posted: Mon Nov 21, 2016 10:28 am
On the basis of your links, there's no way of telling. From knowing what is likely to be inside them, my guess is they'd be fine - but I'd check that with the manufacturer, because some industrial applications specifically need a slow response.

Re: digital steam engine indicator

Posted: Mon Nov 21, 2016 5:22 pm
by fredrosse
These issues are already addressed in this forum thread, April of this year. See previous posts.

Less than one Millisecond response time is easily obtained with electronic transducers. At 500RPM that gives 0.06 seconds per piston stroke, so even the standard electronic transducers could give 60 accurate data points per engine stroke, plenty fast enough to generate excellent engine PV diagrams.

"They also sell a "Pressure snubber for steam and gases" for it." That would not have any application here, it would only make measurements meaningless in all but the slowest engines.