I'm working on a 1919 735b that has had installed rather than the kidney bottle a Bristol Water Level Guage ( pat'd 1920). This is a Weston milliameter and a thermal pile column, connected to the top and bottom of the boiler. I don't know if this is a Derr system since the patent for the Derr seems very different, unless there is a later version. This column has two wires coming out of it at the bottom. The issue is at full boiler pressure the dash guage reads over full at all times, no matter what the water level actully is. I have lowered the column to keep it cooler but to no avail. I don't know if swaping it end for end would make any difference. A resistor could be used to lower the level but then I don't think the whole thing would work properly.
As I understand it there are multiple themalcouples in the column, and when the upper ones get hotter than the lowers it ouputs a higher voltage, so I don't know if turning it upside down would help. Since when cold there is no output; the needle is at LOW, as the temp increases the needle moves to full, so as the temp goes UP when it sees steam it outputs a higher voltage which would tell you the boiler is HIGH rather than low. It seems to me that the guage should park at HIGH and be driven DOWN as it gets hot, or am I full of it. Any thoughts would be most help full (and no he doesn't want the Kidney installed, DON'T go there) Thanks ron Parola

Thermocouple Indicator


The Thermocouple Indicator uses multiple J type thermocouples. A simple single J type thermocouple is two dissimilar metal wires one iron (positive) and one constantan (negative), joined at one pair of their ends, heating the junction creates an electrical potential (approximately proportional to temperature), between their unconnected ends, like a single cell battery, by connecting these ends to a meter this potential can be read.
If we take our single thermocouple and create a second junction (reference junction) by joining the free end of the constantan (negative) wire to another iron (positive) wire we will have two thermocouples (the indicator uses eight), like two single cell batteries, but with one connected backwards, now instead of reading the potential from them both, we will read the difference between them. If we connect the free ends to a meter and heat both the junctions to the same temperature, they will be at the same potential, but opposite in sign and produce no movement on the meter. If we now heat one junction to say 400 degrees and the other to 300 degrees, there will not only be a temperature difference, but a potential difference as well, this will move the meter.
The boiler steam is always hotter than the water, and our indicator has four thermocouples spaced along the water column tube to sense temperature change as the water level changes, and four backwards ones (reference junctions) placed at the top (steam end) of the water column, these will always be hotter than any adjacent to water in the column.
The following simple analogies are not indicative of actual temperatures, design or conditions, and use abbreviated (poor) science as well. (It’s just too tough to do it right)
Low water indication, where the lowest thermocouple is adjacent to water at 200 degrees in the column, and the three above are adjacent to steam at 300 degrees (3 x 300 = 900) (200 + 900 =1100) Our four reference junctions at the top are at 300 degrees (4 x 300 = 1200). The math is the difference between the reference junctions 1200 and temperature measurements 1100 = 100.
Medium water indication, where the lowest two thermocouples are adjacent to water at 200 degrees (2 x 200 = 400) in the column, and the two above are adjacent to steam at 300 degrees (2 x 300 = 600) 400 + 600 = 1000 Our four reference junctions at the top are at 300 degrees (4 x 300 = 1200). The math is the difference between the reference junctions 1200 and temperature measurements 1000 = 200.
High water indication, where the lowest three thermocouples are adjacent to water at 200 degrees (3 x 200 = 600) in the column, and the one above is adjacent to steam at 300 degrees (600 + 300= 900). Our four reference junctions at the top are at 300 degrees (4 x 300 = 1200). The math is the difference between the reference junctions 1200 and temperature measurements 900 = 300.
So in our analogy low is 100, medium is 200 and high is 300; actually it’s the electrical potential that’s proportional to these temperatures that we want, it’s what moves the needle on the meter. But I hope this gives you an idea of how it works.
See attachment
Rolly

Rolly

Hi Ron,

Rolly got there before me but here are my two penny worth!

Thomas Derr's book "The Modern Steam Car" says that he invented the Bristol Derr Boiler water Level Gauge while he was working for the Bristol Company so it's the same thing. There is a picture which shows the wires coming from the bottom end of the column.

The idea is that you have four sets of paired thermocouples connected in opposition to one another. In each pair there is one at the top (Rolly calls it the reference thermocouple) which is always in steam and its twin (the sensor) at some lower point in the column. If both are in steam then the outputs cancel out but if one is in water and therefore at a lower temperature the pair produce a voltage. Since the four sets have their lower twins set at various levels in the column the combined voltage indicates the level of water in the boiler, the higher the level the higher the voltage.

What we haven't worked out to help you is why your unit doesn't give you the right reading. I suspect there must be a faulty thermocouple which fails to balance out its twin so that you are getting a high voltage whatever the level is. Since all the thermocouples are connected in series I guess there must be short across one of the thermocouple ends down the column which lets the current flow but cuts out the voltage produced by that thermocouple end so that pair makes a voltage even when both sensor and reference thermocouples are in steam.

Confusing innit!

Mike

The original Derr unit has a layer of asbestos and string to insulate the thermocouples from the pipe. See attached. When I built mine a used a thin layer ceramic that had a high level of dielectric strength yet very good thermal conductivity.
The thing is they work but have a delay that can get you in trouble. On a hill you can be low on water before the thing lets you know about it.
I like my liquid level probe LED unit. its always on the money.
Rolly

Rolly

Thanks for the explanation from both of you. I did get it working; first lowered the level of the column by about 1.5 inches. The replacement boiler is two inches shorter. And also it seems that since it is a temperature guage rather than a level indicator the temperature sets the maximum deflection. I don't know if the guage is adjustable, there is a screw on the top of the guage itself which doesn't seem to do anything, but if the boiler pressure is higher than the system is calibrated for needle deflection would be farther than the level would indicate (I THINK). So a potentiometer set at about 8 ohms centres the needle at normal boiler pressure AND needle. And Rolly what's all this about an led indicator?? Cheers Ron P

Ron
When I built my Derr boiler for my 1920 I decided I did not want a reflex water gauge sitting between my legs at 600 – 650 PSI. Actual hydro tested for 3000 PSI.
I built a liquid level indicator using commercial liquid level probes, They are not spark, plug and are rated 2500 PSI. They except ¼ - 316 stainless rods, and an LED display. I built the display to look like a reflex water gauge, acceptable in most Stanleys.
See attachments.

See attachments.

Rolly

Ron
The display has five LED lights the center one green, lowest one red, highest one blue. Two lights above and two below for a four inch level.

Rolly

Nice work! That looks really great. Couple of questions. Who manufactures the liquid level probe and what sort of power supply does it use?
Cheers
Mark Stacey

The second electronic water level indicator I built uses a 12 Volt DC power source. It is a simple circuit composed of a high intensity LED and a 1 K resistor. You need a pressure container to hold the number of electric probes required. This is connected to the boiler as any other water column. The probes are wired to the LED indicator panel. This is a very fast acting indicator, and as the water rises and falls in the boiler the LED will light as a bar of lights going up and down. The Probes are commercially available part # WCC-1138. They are good for 2500 PSI. You will need some ¼ inch 303 stainless rod to extend the probes to different lengths. Depending on the pressure or temperature of the boiler you may want to extend the top of the probe for terminating the wires, Teflon coated wire should be used. I used shielded Teflon coax. See the attached drawings and photos. You need to determine the range of height you want to know where the water will be and then how close you want to monitor it. This will determine how many probes you will need. Four to six inches should be sufficient.
The only draw back to this system is the ends of the probes coming in contact with the water over time will build up a corrosion that will have to be cleaned once a year depending how often you use the boiler. It will not stop the system from working though, The water level will only be a little higher each time then what you set the probes for, as the reading will be measured above the corrosion. If you build an AC circuit this would eliminate any corrosion.

Rolly

Nice Rolly,

Good thing is that (if you had room under the hood) you could put the column to the side of the boiler so it is less influenced by steepness of hill, acceleration, braking etc than a standard reflex gauge yet the indicator itself can be in the normal position against the firewall. As the driver of a Stanley with only a reflex gauge and no water level automatic I find the brain is constantly exercised making these adjustments. No time to listen to Steamcar FM.

Mike

Mike you can get a low cost single probe controller at 24V and turn on and off a solenoid buy pass valve to keep your water with in a ½ inch of where you want it.
One probe in a one inch water column will do the trick.
[www.ssac.com]

What are you doing with a reflex gauge anyway, Stanley never used one.
There not even approved for over 300 PSI without a mica sheet and flat glass in them.

Rolly

I once had a reflex gauge glass blow at 450 pounds. Due to the small 5/16" tubing steam lines going to the relex gauge, there wasn't any more against my legs than if some one had thrown a cup of vegtable soup against them. The blow down through the gauge that followed was more of an experience. It was like standing in front of your Stanley during blowing down the boiler. Again, no discomfort but a very irritating inconvenience to have to replace the gauge glass. The cause of the broken gauge glass was because of a warped gauge body that the gauge glass rests against. When replacing the gauge gaskets in a reflex gauge, always check the surfaces with a straight edge. The small 5/16" tubing lines to the reflex gauge are what makes the reflex gauge safer to use. In a locomotive, typically, much larger and more dangerous 3/8" to 1/2" pipe is used to plumb in the reflex boiler gauge. With no moving parts and no electricity to rely on, the reflex gauge is my first choice of a boiler gauge to use. The last reflex gauge that I purchased for one of our Stanleys was a Penberthy model 1RM-5. It is rated for use up to 1950 pounds at 600 degres F. Reflex gauges for steam boats are usually rated to only 300 PSI.

Earst Reflex 545 gauge glass is rated 500 PSI at 450F for use with saturated steam.
If you talk with Earst they don’t recommend going over 300 PSI for steam.

There flat glass with mica behind it is rated for saturated steam, you can get to 608F at 1000 PSI for 546 flat glass and 1500 PSI for 547 Flat glass.

Rolly

I had bought a new steam boat boiler level reflex gauge two years ago that was pressure tested at 600 psi cold water. It was rated for use at 350 pounds steam pressure. The lower price tag was attractive. On our Mt. Wagon at 550 psi steam pressure, it warped the housing like a potatoe chip and it blew it's gaskets. The manufacturer took it back. For fewer problems in the future, I found that it is wise to use a reflex gauge that is rated twice the working pressure that you are using it at. It is not so much about the danger of blowing the reflex gauge as it is more about the reliability of the gauge and not being broken down along the side of the road because of it.

Does any one know what they used as a water gauge on the 06 and 07 racer(beach)? Wondering how they countered the higher pressures for the runs.

The VCR's I think lived at normal pressure so they shouldn't have been a problem.

Caleb Ramsby

Yes Rolly - it's hard to find a manufacturer who will supply one with a 600psi rating but they do work.

Like Sssteamer I have 5/16s pipes but have also fitted a an emergency restrictor on each connection which has a stainless ball in it which passes slow flow but flies up and blocks most of the flow if the glass burst. I found I had to put a slight nick across the seat with a file so the ball would release if I went too hard with the gauge blowdown. I carry a spare glass and fit a new one every couple of years. Never had one blow. Hardest thing is to see the level on a sunny day even with a light fitted but I make sure I know where it is and can give a quick blow down when driving to drop the level and be sure to see where it comes back to.

Mike

Does any one know what they used as a water gauge on the 06 and 07 racer(beach)?

Caleb
Up until 1909 Stanley used the bucket type water indicator. Some where between 1909 and 1910 they switched over to the three tube indicator, and latter to a newer version of the three tube indicator, and finishing up with the kidney type.

I was wondering about the same question. I took a bucket type all apart to see how it worked and do not believe it would stand up to the reported pressure the racer used.

The reflex type gauge was available at the turn of the century and I believe was used on river steamboats in the late 1890’S. The question is did Stanley ever use one on the racer.

My gut feeling is he ran with out one, if the boiler was full to its operating level he could not use enough water in the run to get low enough to hurt the boiler, and most likely killed the fire toward the end of the run.
See attachment

Rolly

The reflex type gauge was available at the turn of the century and I believe was used on river steamboats in the late 1890’S. The question is did Stanley ever use one on the racer.

I was wrong, the reason Stanley did not use the water glass, as we know it, is because it was invented by George Whitney patented in 1908

Stanley had goon through one lawsuit with Whitney by this time.
See attachment

Caleb Thanks for asking the question, I learned something.

Rolly



Edited 1 time(s). Last edit at 05/14/2008 10:26AM by Rolly.

Rolly,

I was thinking in the same direction, as to the them not using a gauge in the runs with the beach car. I remember hearing that they made the runs with the pumps off, so that they wouldn't break them. They may have used a column and a few gauge cocks to see where the water level was before they fired up the car and made their run.

Great information about Whitney and his glass gauge that you discovered.

Thanks for your answer to my question, I learned something too.

Caleb Ramsby