There are 2 possibilities left: - find the formula; - take advantage of the fact that you have SW to simulate the diameters corresponding to the graduations you want and take the corresponding graduation position.
Already in terms of precision there is something that catches my attention because it will be worse with real mechanical parts.
You have no less than 7 axes in your joints so with the more or less cumulative sets and especially with the largest connecting rod you can't have a reliable measurement because you have to have a multiplier coefficient of the play which is due to the large connecting rod.
I don't know if this "phimêtre" exists or if you came up with it, but it seems to me that a simpler system should give you better results.
It all depends:
- reliability during repeatability (do you always measure the same diameter you get the same result - the mastery of the cumulative play of the axes - the trigonometric calculation that you use or not - The precision sought because, for example, if it is to be used on a bending machine, precision is essential.
In fact it depends on the use you want to make of your "phimêtre" in any case it will not be for metrology :-)
The purpose of this tool is not to do metrology, so an accuracy of +or- 3mm should be enough. It is a question of measuring pylons to adapt accessories by strapping. I am reassured to learn that the graduations cannot be equidistant in this type of measurement, this will allow me to redraw them by calibration with different tube Ø.
I don't know if such a "phymeter" exists but a digital model sells for 320€. So I started with a cheap model. Will it work properly?
@zozo_mp. Talk about a simpler model. What are you thinking about?
Yes, Stefbeno is right, in fact a room that looks like an inverted T allows you to instantly know the radius of your room.
On the leg of the T you have a simple ring that slides and will instantly give you the arrow thanks to a simple graduation in mm.
The horizontal bar, whose length is fixed (or known if it slides), gives you the string of your arc.
The ring that slides tangles the top of the circle. At zero (flat surface) you have a string but no arrow. The more your arrow increases, the more your ring moves away from the rope and thus gives the length of the arrow.
Anyway, you always know the arrow and the rope to know the radius you have to
You do the calculations for the most common spokes in your factory, you engrave them on the leg of the T and this simple system gives you a direct reading on your T with Slip Ring
@gt22: you made a mistake by a 0: 1m=1000mm (and not 10.000). Nevertheless, an amplification seems necessary, see below
For something that can be handled, I will start with a 500mm rope base (to be able to be held with both hands without being quartered and to have a correct measuring base).
With this base of 500 and a Ø1000, the deflection is 66.99mm, with Ø1003 we go to 66.76 or 0.23mm. There is a need for amplification: rather than connecting rods, gears can be used to drive a finger either parallel to the probe or via a toothed sector. Given the necessary amplification, I would prefer the second. A small spring to keep the probe in contact, possibly a locking/unblocking system by the operator to keep the measurement.
We know the required accuracy (+/-3mm over 1000mm), We should know the measuring range (mini-max). If this range is too large, it is always possible to make a double graduation for example or to provide interchangeable graduated sectors (but beware of the risk of reading error...)
I thought a little about your problem and I wondered, to measure large diameters wouldn't it be more reasonable to use a tape measure?
By that I mean that it already exists and that it probably needs to be standardized. After all, I'm not against the creation of a new tool, it's always interesting:)
The "string depth gauge" system already exists, but in the digital version and indeed in the mechanical version, the variations are so small that it is difficult to have a reliable measurement. Before coming to see your opinions, I was thinking about gears. I'll look into that, especially since I've never used gear or rack constraints, but anyway, in either case, you need a gear ratio.
You will find attached a schematic representation of my idea.
Starting from the principle that 3 points (A,B,C) belonging to a circle form a triangle inscribed in this circle, we can use the law of sines to calculate the diameter of this circle.
By knowing the RD1 dimension and doing a little trigo, we can calculate the position of the point B and therefore calculate the distance AB. Then, a little more trigo to calculate the RD2 distance. Once RD2 has been determined, all that remains is to apply the law of sines.
Well all this to say that you just have to add a protractor giving the diameter directly on one of the arms of the compass either by modeling or by calculation...
@Yves.T: It's better to put the image in PJ, you can't read anything on an integrated image. Interesting idea, we only have rotating guides, there remains the problem of reading. For a Ø1000mm, 300 arms (RD1?), the angle is 146.80°, for a Ø1004 it goes to 146.93°. You have to seriously amplify the difference in order to be able to read it.
Indeed, it's that you can't put two attachments and having posted a message before, I didn't want there to be 3 or 4 in a row just for the attachments.
Yes, you have to amplify but with a protractor of about 200 mm and a caliper style vernier there should be no problem.
By knowing the RD1 dimension and doing a little trigo, we can calculate the position of the point B and therefore calculate the distance AB. Then, a little more trigo to calculate the RD2 distance. Once RD2 has been determined, all that remains is to apply the law of sines.