I have at various times tried to check three local subdivision plats from the 1950??s and have yet to find one that I can get to fit with themselves at a reasonable accuracy (never mind what??s on the ground). Is this the ??normal? case?
I??ve been fighting one and am so frustrated I need to vent. I pulled the plat from the on-line records for a friend and computed chords for his frontage curve so he could measure to look for monuments, in preparation for rebuilding a stone wall. It turns out there are monuments, but his frontage and the adjacent ones all measure a foot or three in excess of the plat dimensions (7.7 ft excess in 178 or 4.3%), which has us scratching our heads. If there are discrepancies, I??m not surprised they are on a curve.
So to start I computed the exterior of the 20-lot subdivision. Angles given to 1 minute sum just 1 minute off, but the closure is over 2 ft. Least squares holding the given angles tight wants one distance to increase by 2.0 ft north-south (43.57 to 45.57) and another to shrink from 106.95 to maybe 106.095 NW-SE. Those changes give excellent closure of the exterior, but I can??t get the lots to fit. There is 2 or 3 ft floating around when you try to fit all the lots in, either with or without changes to the exterior. One curve checks out with its parameters pretty well. Others don??t fit within many tenths to a foot. I??ve tried looking for likely misreadings of the numbers on an imperfect photo of the plat, tried holding or freeing the delta angles and/or radii, holding one area tight to see what others fit, etc.?ÿ The city GIS shows the same numbers I read from the plat, but their distances don't scale out to match the labels.?ÿ Nothing gels. There aren??t enough identifiable points (fence lines, etc.) on Google Earth to use occupation to look for misreadings at the 1 or 2 ft level. I??m pretty much at an impasse.
On another 1950??s plat I once looked at, the exterior boundary angles given failed to close by 42 minutes. Playing with holding various interior lines failed to identify a simple copy error (or two) that would allow things to fit well.
There is another 1958 plat with an exterior that doesn??t close by about 10 feet. I think the bearing of the south line was computed from the center of the street, forgetting that the street being extended wasn??t centered on the quarter-section line, so that one is easy, but I??m not sure how much that affected the interior. It has some lots with their rear corners offset from each other, and the distances on each side of the common line add up 0.8 ft different. It has right angles indicated where a lot dimension would have to change 0.5 ft to match. Some of the curve data given is self-consistent with arc formulas, others almost match chord formulas, and some just don??t make sense. Trying to fit anything around the curves doesn??t seem to work with at least a foot either missing here or left over there.
So my question: did I just stumble across the bad ones, don??t I have the right technique for finding scrivener errors and misreadings, is it an indication of poor local practice at the time, or is that just the way things were everywhere in the 1950??s?
I just did a retracement on a subdivision lot from 1960 on Friday and it was so lacking in the basics that it was not able to be plotted without making some huge assumptions and taking guesses. I always key in the area where I am working for a plot sheet to take in the field but I could not do so on this as there were just a couple of bearings, no readable angles, no curve data, distances seemed to be a suggestion. This was possibly the worst I have seen and signed by a engineer. First time I had been in that subdivision and the last.
In the areas of California that I'm familiar with, 1950s-era plats are pretty uniformly tight, with misclosures of more than a few hundredths of a foot in any closed figure being uncommon.
Around here we see the same stuff. Usually there isnt enough info, couple of the engineers around here were notorious for these.
Some of the ancient surveyors of that era were still using a compass.
Others were using no instrument at all and were merely cross chaining.
I followed one that was in a magnificent pine tree farm with large trees at the time and many were in the way and straight lines on paper were meander lines thru pine trees on the ground.
Anything withing 3ft of location was normal and being as much as 5 degrees in bearing was expected for some boundaries as the magnetic deviations were caused by many factors, probably tools in peoples hands not knowing any better.
Some days the math, paper descriptions and distances or relative common sense does not apply to the location of a found monument.
Some.?ÿ Some are not.
Did anyone ever try out Dennis Drumm's Primacode Transform for Windows software?
His fundamental believe was that the angles were better than the distances.?ÿ Input the record coordinates, input the surveyed coordinates and it comes up with a solution by scaling the record plan.?ÿ While this will not resolve your angular discrepancies, it provides an insight into remedying the measurements of yesterday.
Some.?ÿ Some are not.
Did anyone ever try out Dennis Drumm's Primacode Transform for Windows software?
His fundamental believe was that the angles were better than the distances.?ÿ Input the record coordinates, input the surveyed coordinates and it comes up with a solution by scaling the record plan.?ÿ While this will not resolve your angular discrepancies, it provides an insight into remedying the measurements of yesterday.
That transformation finds a rotation, a scaling, and XY offsets for a best match between plat and measured.?ÿ It sounds like a good approach when you have full measured data to work with.?ÿ It also has the ability to pick which things to hold and fit the rest.
I've been tempted to try to write a simple version of that software myself, but haven't tackled it.?ÿ I did do my own 2-D plane LS software (checked against Star*Net demo on simple examples) that I'm using for the plat analyses.
The OP was about just figuring out a best fit to the plat given only what is written on it.
Lot 57 was the one I did. It took all day but I got it.
?ÿ
I don't have many from that era, but they are a pleasure to work with. Sorry to hear you have a different experience.?ÿ
As with today's plats the quality varied wildly. In the 1950's, unless you were in a heavily urbanized area there were not may people who cared if a plat was a foot or two off.?ÿ Frankly, I don't loose much sleep over it on modern plats either. The corners are where they are. Small misclosures don't matter at all unless you have to resort to proportioning, an then, as long as you choose an equitable method, the problem will be solved.?ÿ?ÿ
There are much worse ones out there. When you cant even figure out how to check the closure is when you have real problems.?ÿ
?ÿ
Hey Bill,
If these are in CR, I know of more than one subdivision from that era where the interior lot lines were not surveyed or calculated. They were probably scaled. It appears they probably did a random traverse, calculated the boundary and did monument the exterior corners for the most part, but the accuracies are problematic. Locating the interior corners is obviously a challenge.?ÿ
?ÿ
Don, good to hear from you! Welcome to the forum. I see you have been lurking for a while but this was your first post.?ÿ Are you still in the QC?
I recall talking to you some about the subdivision the church is in, which is the 3rd one mentioned in the post that started the thread.
I would check the curve data to see if it is based on a chord/arc definition and try the opposite, or look for a spot where a small/short tangent were left off of the drawing before I chose to recalculate the whole plat. The 50's were boom years when many individuals were thrust into jobs that they had little or crossover experience with. The Curta II was introduced to this industry which might have caused a systematic operator error; if you can isolate it. I also have had luck getting a copy of the same plat from another source and found missing or transposed information. Street widths are not always what they seem to be.?ÿ
But than you probably already thought of all that.?ÿ
This is an example of the ones I see that are almost impossible to resurvey. But this one is 1910, the 1940-1960 ones are really good.?ÿ
?ÿ
Bill,
Yep, still in the QC, keeping on keeping on. ?ÿ
My take on some of the 50's, 60's & 70's work follows our profession's somewhat rocky road post WWII when a lot of civil grads easily obtained their LS license without the education and experience that was a big part of Civil Engineer's careers prior.?ÿ
Some of the best work I have seen was from a lot of the earlier work (not all of it). There has been improvement over the last 20-30 years, but we all know of some sloppy work still being done. Here's to hoping we get some younger folks involved to shore up the practice.
?ÿ
?ÿ
I've found most of them are, but not so you could do a best fit on the computer in many cases.?ÿ I can tell by the surveyor now in my area.?ÿ Some will be right on, others 5'+- (random directions) of plat that doesn't close.?ÿ If things are fitting within 5 ft. and find something 10 ft. off, then have to consider it might be disturbed or not a surveyed position.?ÿ Sometimes even the good ones had trouble with curves though.?ÿ My guess is they drew them in and could then measure a radius and chord, but the curve wasn't tangent.
The task at hand on these is to place the error where it occurred. That will require gathering the evidence on the ground. If I can't see a reasonable source of the problem before going out I don't bother pre-adjusting the data.
This is where understanding the equipment and methods of the past really helps. We don't need to do a senester of verniers and chains, but we absolutely must give the basics...
The 1850's plat of the "City of Portland" is remarkably self-consistent. But there aren't any curves in it. And the oldest monuments you can hope to find were set 60 years after the plat was filed.?ÿ?ÿ