Hi all,
?ÿ
Long time reader first time poster. Any help will be greatly appreciated.
?ÿ
I am being asked more and more to undertake structural movement monitoring for domestic basement construction. I was wondering if anyone can shine some light on their recommended processes for establishing and maintaining a reliable control network.
?ÿ
I have tried a number of techniques and each has brought differing results. I've found that working to sub-millimetre accuracy doesn't allow for the use of nails as my permanent/primary control due to the inaccuracies in holding a mini prism. I will always set-up with a resection and never over a station for this reason. Generally my set-up results for a resection using targets will be 0.5mm or less in E, N & H, and usually 1" or less in orientation.
However, whilst retro targets are ideal, I am concerned that they do not provide the longevity of a permanent control network. I know monitoring works are only ever temporary (at least in my applications), but I am concerned that if by chance all of my targets were to be picked off, it'd be game over. I've tried to use reflectorless points (paint marks on buildings for example) but centring over these is not accurate enough for my liking due to the lack of cross hair/defined point. It is common for me to (cheekily I know) stick retros on peoples property without their permission - usually low down on garden walls and in discreet places. I will also use items such as the base of bollards and lamp posts and actually it is extremely rare that people remove these/notice them.
I guess my question is: do you guys utilise both primary and secondary control networks for small scale domestic monitoring surveys? Or do you entirely rely upon targets in reliable positions and on the basis of good geometry and consistency?
?ÿ
Really appreciate your time.
Chris
Each site has its own challenges. Length of time for the monitoring has a big influence on the quantity and quality of control that I set. When I perform monitoring on mountain slopes, I will be sure to place reflecterless targets well away from the AOI.
Establishing realistic monitoring tolerances is the first and most important step. It's up to you to explain that a tolerance requirement of ??Any movement? is unrealistic.
Using a high quality level and Invar rod will give better and faster results if elevation is your main concern.
Using two different colors of reflectorless targets, both cut in 1.5? squares, I put a 1/8? hole in the center of one and then stick it on the other which creates a color contrast that I can site. Sometimes I stick them on aluminum flat stock that I then drill and fasten to walls or other solid structures.
Like others here, I take advantage of Star*Net and use it to calculate the amount of redundant reading required to meet my minimum tolerance. A tightly run control network works, but I prefer resecting to reflectorless targets now that I've experimented with both.
Sub-millimeter isn't really practical, or required in most cases. I mean, it is going to be super expensive to achieve and typically structures will move that much, and more, in the normal course of a day.?ÿ Architects writing specs may state that they want it that close without know WTF they are asking for. I do monitoring regularly. I tell my clients that until the numbers show things moving at least 0.02', and staying there, they haven't moved at all.?ÿ ?ÿ ?ÿ
Thank you for taking the time to reply.
We operate to an absolute accuracy of +/-1mm for readings, but generally I am aiming to ensure my resection set-up is 0.5mm or less in position and elevation in order to set a standard for each job/survey. We report readings to the nearest millimetre however. A typical trigger value for our clients will be 8mm before works need to stop (not like they ever do though!).
I think going forward I will rely on a single but tested network of targets only and ensure regular control checks are undertaken. I have not used Star*Net before but I have noticed many people talking about it and may give it a go.
Ultimately what I am hoping to determine, based on more experienced surveyors opinions, is a practical but reliable system for establishing a form of control which yields reliable and consistent results. I have concerns that retros can be picked off and if I lose a certain number of these I am in trouble. I have recently invested in some 10mm square retros (which are tiny!) in order to keep my control points discreet. I haven't used these yet but I am hoping they will be less noticeable and therefore likely to remain in place.
Do you have any advice for maximising survey accuracy for recording base readings and observing these each week? For example, do you swear by undertaking observations in both faces to take an average? If monitoring for X, Y & Z displacement, will you survey for X & Y with a total station using a resection and then follow up by undertaking a height transfer to determine the Z values (assuming a digital level and invar aren't feasible)?
Thanks again for your help!
?ÿ
?ÿ
?ÿ
If you are already achieving submillimeter accuracy on your resections I would continue doing what you are doing.
I tend to worry more about the stability of my control. Everything expands and contracts. Precise measuring is less of a concern than being aware of the influences of the environment on my measurements.
I had a long conversation with a NGS surveyor and picked his brain about the various ways to establish long term control points that are not affected by the weather. Long story short, there's a valid reason that NGS is moving away from physical benchmarks.
If you want to get better than .5mm you may need to take all your shots late at night within a narrow range of relative humidity.
Everything expands and contracts.
And that includes buildings and their foundations, when you get down to that kind of precision.?ÿ
There's a reason iron rebar works so well in concrete - the coefficient of thermal expansion for iron, concrete, and some stone are all somewhat similar.
Generally we always achieve submillimeter resections and this is because we test them over and over at the time of installation/monitoring set-up.
You are right regarding general movement. At the end of the day I suspect monitoring by it's very nature is subject to accuracy restraints and this trickles all the way down to my control, no matter how effective it is to begin with.
To an outsider it probably appears that I am overthinking things, but it's because I am very determined to provide my clients with ultimate confidence in our results and ensure longevity in how we operate on each job. This forum is great for gaining insights into operations and recommendations from other surveyors.
Thanks once again for your valuable contributions. I appreciate your time and effort.
do you swear by undertaking observations in both faces to take an average?
Yes, also the best way to check your instrument and determine somethings not OK.
I once monitored a bridge that got hit by the 2009 storm in Georgia with a two-lane highway.?ÿ?ÿ I began in May 2012 and ended in February 2013 with 7 observations over 10 points that I drilled and glued into the bridge. I used a Leica DNA10 Digital Level and the most movement I saw over that time period was 0.026 feet. It was a steel and concrete structure and that is just the most extreme movement that I noticed over all the points.?ÿ
I have almost 20 years of annual monitoring on several projects. I can say with certainty that parts of millimeters isn't happening without uncommon practices. That may be what the data collector says, but a valid network analysis will undoubtedly reveal uncertainties several times that.
The key to good monitoring (after sound network design) is to screw things up the same way every single time. Trying to improve after you started introduces different errors, with near certainty of differeing results.?ÿ
Get a copy of Kuang on network design, and follow up with procedure research for the measurements.?ÿ
Secondary Control outside of the monitoring area is a must. Small mag nails set flush in concrete work well, just get a small mini prism point (Leica Mini) just for that job.?ÿ
I can say with certainty that parts of millimeters isn't happening without uncommon practices. That may be what the data collector says, but a valid network analysis will undoubtedly reveal uncertainties several times that.
What he said ????ÿ
If you want to measure for a living: Buy it..Read it...Live it
?ÿ
The data collector may say this or that, but bring that data back to the office and run a least squares analysis and see what your error ellipse are.
?ÿ
Take a Leica Viva TS16, the published accuracy is 1mm+1.5ppm. Centering accuracy on a Leica tribrach: 0.5mm. Leica reflector GPH1P Centering Accuracy 0.3mm (the others are between 1mm to 2mm. Leica's best tripods contribute up to 2" of error. Add it all up. Plus the bleary eyes of the instrument man, plus wind, plus varying weather and atmospheric conditions...sub mm isn't really in the capability of my tools.?ÿ
?ÿ
I'd never seen that book before - it's viewable on line for free, and looks really good - used copies not expensive. I'll be picking one up. Thanks!
One of my Grandson's favorite train crashes!
I made him a wooden train and he reenacted the crash numberous times.
You certainly get no argument from me regarding errors.
Unless you invest in the right equipment and adhere to strict procedures, you aren't cracking millimeters.
From what I understand it’s pretty much the standard text on the subject for laboratory sciences. On the top shelf
We have the same edition of Clark's on our shelves. We also have identical shelves in our offices.