tips to start a verification of a mechanical structure

[Powermos]

Bye to all,
I am new to the forum and I am delighted in the mechanics by hobby since by profession I am an electronic ing.

I'm always building a four-fold solar biaxial chaser.
using rhinoceros I created the 3d of the whole structure, I calculated the action of the wind on the surface of the panels, sized the support pole using the resistance module and the type of material and the various bolted unions (I am seeing the calculation and verification of welded unions collecting material to the right and left).

Now I would like to do a numerical analysis to have comfort of the calculations that I have done (as already said I am not an strutturist, I only read several dispenses found on the net) and therefore have a reasonable security that once fitted the system all does not present structural yields (at first the safety, according to the cost since I will have to make realize mechanical details with dimensions of a certain relief from specialized workshops).

for the reasons mentioned above I allowed myself to disturb to ask if someone could recommend an analysis software to the finite elements, possibly free or functional demo for at least 30 days (so as to allow me to do some verification), for structures composed of poles, pipes, plates and other elements like hinges etc... and that allow me to assign the external forces and assign the materials and geometries of the individual details to determine the stresses on my structure so as to see if the maximum dimensioning that I have done is correct. If then geometries could be imported as iges I would also be halfway through the work since I would only export the various groups directly from rhinoceros.

for this purpose, to give an idea of the structure that I should analyze and then realize in practice, I attach a set of screenshots taken from rhinoceros.

since I think that the vision of images does not provide a precise idea of the size of the structure I add some details:

- the support frame of the panels consists of profiles in steel with square section thickness 1.5 mm sadati each other and all supported by the three plates in fe360 from 110 mm wide for 10 mm thick that form the central h of support of the frame. the overall dimensions of the frame are 3.6 m long for a height of 1.6 m. the complete frame of photovoltaic modules, brackets and drawn has a total weight of about 110.64 kg (theoretical weights calculated with the tables of the materials). the three h plates are welded between them and the central plate and bolted through 12 m12 bolts to three snode hinges (of those used for the 100 mm oleopneumatic cylinders) and are used as snodes to tilt the frame respect the horizon.
- the support pole is a fe360 (s235jr) with external diameter 127 mm thickness 8 mm, total height of about 896 mm.
- the base is a square frame that I use essentially to do preliminary tests, for the definitive installation I think I will make a concrete base to prevent the wind if you take it away.

are obviously well-liked suggestions and advice from you mechanical and strutturist experts since you will certainly be able to assess whether my idea was conceived correctly or if there are any underlying vices to correct.

are however welcome examples and the help of chiuque wanted to spend a little time (I know that it is very valuable for all) to help me in this experience (I regret not having entered calculation tests of the structures in the course of graduation, it is a fascinating topic, but I had already several and also I never thought that one day I would have come across in the realization of a mechanical system of this kind...).

thank you in advance for the attention you will want and you will be able to dedicate me.

greetings
♪

 

[myface]

hi, you deserve an answer even for how you expressed yourself and for your aversion to pressapochism (weight about about 110.64 kg he:smile that many technicians should possess.

before facing the calculations, I wanted to comment on the system, which is improved in the area of the actuator that commands the inclination: so as the support seems to me "light", especially with respect to the possible twist, and it seems to me also very long. Take them as impressions, of course.

what inclination range should your tracker meet?

 

[meccanicamg]

Hello powermos,
Nice work and welcome. free software for finite element analysis I don't know that they run under wins. but under linux there are different ones and if you try to search in the forum you will find some proposals.
Remember that the values you will find from the fem you can not take them for 100% good as they depend a lot on the characteristics of the constraints and forces applied, from the points of action and also from the meshatura. You could find singular points and it is not said that they are real. Therefore you always believe the calculation values according to rules and rules of good technique.

Good job

 

[Powermos]

hi, you deserve an answer even for how you expressed yourself and for your aversion to pressapochism (weight about about 110.64 kg he:smile that many technicians should possess.

before facing the calculations, I wanted to comment on the system, which is improved in the area of the actuator that commands the inclination: so as the support seems to me "light", especially with respect to the possible twist, and it seems to me also very long. Take them as impressions, of course.

what inclination range should your tracker meet?

Hi.
thanks for the attention, I try to deepen the description of the structure, perhaps providing some general details about what is a biaxial and monoaxial solar pursuer.
the chaser with regard to the azimuth rotation (in the horizontal plane) is equipped with a raddle with gear motors, this system allows you to make virtually a complete turn angle, in practice it will always be to rotate within an angle that is always understood within 270° (which develops from sunrise to east until sunset to west).
this rotation is commonly indicated with the daily term (daily) since it is the one that allows to chase the apparent motion of the sun during the day.
the linear actuator allows the positioning of the modules on the horizon (tilt angle) and generally this movement is called seasonal since it varies, in fact, predominantly with the seasons.
this means that in the case of a single-axial chaser, for our latitudes and longitudes, you can fix an inclination regarding the horizon of about 30° and rotate the modules along this inclined axis, possibly to maximise the yield even in winter periods you can make manual adjustments increasing the inclination of the modules up to reach the 40 ° regarding the horizon.

returning to the object of desire, I think that you (I allow you to give yourself some...), refer to the tilted bracket that starts from the structure and ends on the hinge of the actuator.

the bracket is welded to the fitting that above carries the plate to which the three hinges of the frame snode are bolted (the support plate has a length of 1000 mm, width of 160 mm and thickness 12 mm, welded to the central connection with two wings of reinforcement) and below is bolted to the azimuth rotation raddle.

this bracket has a thickness of 10 mm and is 50 mm high, the hole for the fixing bolt of the linear actuator has a diameter of 13 mm, while the outer radius of the bracket is 25 mm, this means that I have a circular crown of 18.5 mm around the hole.

the total weight of the linear actuator is about 10 kg and the center of gravity is practically in the snode point (individual using the functions of rhinoceros "mass properties"->"volume centroid").

from a first verification made following the information found on this link:
http://dida.fauser.edu/dispro/carbonar/collegamenti/framecollegamenti.htmI have seen that using always fe360 material and for sigmar (as written in the link of the notes mentioned above) using the value of the sigma amm = 160 n/mmq both the break for hairstyle and for tear of the bracket should be averted, as written before the actuator has a weight of 10 kg in static operation; in the dynamic case I have to win the mechanical frictions of the first detachment of the hinges and the moment of inertia of the frame with the panels in order to put in rotation the frame itself. even here I had made some accounts that I have not now underhand, but it seemed to me that it was all right (thinking well maybe I had only considered as a component to win in the worst case, with vertical panels, only the couple of rotation associated with the frame and the modules and perhaps neglected the moment of inertia of the structure since the angular acceleration at the end is very low).

Do you think the calculation methodology I used is good or do you recommend doing some other kind of verification/consideration?

one thing I would like to check is the welding of the support bracket of the linear actuator at the top chord, so at first glance I sized things by providing a welding cord from at least 7 mm, can you give me some advice on how to make the calculation/check of the welding and some dispensers to be able to study me before I can make an idea of what to do?

for the hole in which I will block the actuator I had also thought about the use of a drilling compass with din 172, the one with 13 mm hole has a 26 mm collar and 18 mm outer diameter, if on the one hand increases the strength of the hole on the other reduces the crown of material that remains around the compass, considering then that the movement of the actuator is very slow and does not leave the fore

I attach an image of the detail of the group to which the actuator bracket is attached.

thanks for the availability shown.

Hi.
♪

 

[Powermos]

Hello powermos,
Nice work and welcome. free software for finite element analysis I don't know that they run under wins. but under linux there are different ones and if you try to search in the forum you will find some proposals.
Remember that the values you will find from the fem you can not take them for 100% good as they depend a lot on the characteristics of the constraints and forces applied, from the points of action and also from the meshatura. You could find singular points and it is not said that they are real. Therefore you always believe the calculation values according to rules and rules of good technique.

Good job

Hi.
thanks for the directions provided, I had done a preliminary search in the forum, but I had not identified a specific software. I try to see if I find something under linux, thank you.

greetings
♪

 

[cacciatorino]

hi.
thanks for the directions provided, i had done a preliminary search in the forum, but i had not identified a specific software. i try to see if i find something under linux, thank you.

greetings
♪

www.caelinux.com find a packaged version of pre-post processor and solutor fem. then externally there are calculix and elmer that i never used. i think these three are the most popular.

consider that learning to do reliable fem analysis takes a lot of time and effort, so maybe at first you will find yourself using analytical calculations to check fem analysis, rather than the contrary! :smile:

 

[Powermos]

place below some considerations that I have made (repeat, as profane in matter) about the verification of bolts that must support the frame.
the frame is bracketed to the hinges through 12 bolts (3 hinges, those that are used for the 100 oleopneumatic cylinders), but to save I thought to use only two hinges, so I will check it by hypothesize to have only 8 bolts that must support the overall weight of the frame.
in the case of vertical frame compared to the horizon all the weight of the frame weighs on the bolts. in this case wanting to dimensional the whole so that the bolts work for friction I have to verify that, with the precarious couple given to bolts and the friction coefficient between the hinges and the bracket, the resulting friction force is greater than the weight exercised by the structure on the single bolt.

the overall weight of the frame is about 111 kg, if a perfect symmetry is hypothesized then I can also assume that each bolt should be charged 1/8 of the total force or in the verification report:

(fi-n)

the value of the cutting force t associated with the frame that tries to make it flow away from the hinges will be equal to:

t=(frame size * 9.81)/8 = 136.11 n (2)

in (1) fi is the internal traction force resulting from the bolt tightening couple.
speculating to use 8.8 class m12 bolts (the hinges have four 14 mm holes) the clamping pair should be about 89 nm, this produces a traction force inside the bolt of 37 kn.
in the case of the frame placed in a vertical position, the presence of an external force of traction n that acts in the sense of separating the frame plate from the hinges.
to be on the side of the reason I hypothesize that the friction coefficient is equal to 0.3 (not particularly treated surfaces).
under these assumptions I get:

7400

that is, assuming valid the assumptions above, 8 bolts m12 class 8.8 should be able to support for friction the weight of the frame.

At this point I ask the opinion of those who are much more experienced than me in the matter to know whether the account I have made is too simplistic or if it can actually give me an idea about the fact that the structure should remain still on.
wishing to increase the level of detail you could also consider the effect of snow that settles on the top of the frame in vertical, but in this case it is only about increasing the value of t in a congruous way with the tables provided by the various calculation regulations, that unfortunately I do not have..., but I was interested to know if there are macroscopic errors in reasoning or if they are on the right path.

thanks as usual of the time you will devote to me.

greetings
♪

 

[meccanicamg]

Hey, power,
for verification of welded joints and bolted searches in the design forum that we have done some exercises and there are several useful attachments and normative remakes.

If you do a google search you will find university dispenses on bolted and welded unions according to the old regulations collected a cnr 10011 and according to that new eurocode 3.

I remind you that for the ex fe360 it is considered as admissible sigma 160 mpa for laminate thicknesses less than 40 mm. for tangible actions the admissible tau is the octaedral one therefore tau adm = 0.58*sigma amm = 92.5 mpa.

with regard to welding, check the class joint ii or with full penetration or angle cord depending on the execution at least according to the old one cnr 10011. I remember you find tips and exercises to understand how to do.

with regard to the screws m12-8.8 in hole ø14 if only to cut category c is:
fs,rd = ks*n*fpr/1.25 = 9.6 kn
con fpr = 0.7*as*fub = 42208 n = 42.2 kn
with fa = 0.3 brushed surfaces to the flame without rust
fub = 800 mpa load break 8.8
as = 84.3 mmq
ks = 0.85 hole increased because m12 normal is ø13 mm
n = number of sliding surfaces = 1

fs,rd is the limit to compare to the cutting design force for each bolt at the last limit state. then you have to make 2 more verifications of which one to recroach and a plastic resistance test in the holes section. however evaluates if esite only cut or combined cut and traction. the moments in the structure cause traction. so do not fossilize you on the formula I wrote before but follow scrupulous verification according to regulations.

for friction verification and tightening pairs see ex uni cnr 10011 and/or eurocode 3(en 1993-1-8:2005). it is recommended in the execution of tightening with dynamometric key if you want to ensure a known pair.

 

[Powermos]

@meccanicamg

Thanks for the calculations and information provided, I was lucky and I found the cnr standard on the net, but I couldn't find details about eurocode 3 (Eurocode 3(en 1993-1-8:2005).

I try to orientate myself in the meanders of the cnr norm and to make the necessary checks.

Thanks again

greetings
♪

 

[meccanicamg]

@meccanicamg

Thanks for the calculations and information provided, I was lucky and I found the cnr standard on the net, but I couldn't find details about eurocode 3 (Eurocode 3(en 1993-1-8:2005).

I try to orientate myself in the meanders of the cnr norm and to make the necessary checks.

Thanks again

greetings
♪

Okay. we say that the cnr is fine and unlike the ec3 is more tolerant, so if you find a security factor 10 ....ok but if you are close to lower values, do not hesitate to oversize.

 

[Powermos]

with regard to the screws m12-8.8 in hole ø14 if only to cut category c is:
fs,rd = ks*n*fpr/1.25 = 9.6 kn
con fpr = 0.7*as*fub = 42208 n = 42.2 kn
with fa = 0.3 brushed surfaces to the flame without rust
fub = 800 mpa load break 8.8
as = 84.3 mmq
ks = 0.85 hole increased because m12 normal is ø13 mm
n = number of sliding surfaces = 1

By reading the 1993-1-1 (only I was able to see) on page 151 I find the formula that you indicated above, I find all coefficients except the value of the partial safety coefficient that you placed at 1.25 while I would put 1.4 since in 6.5.8.1(4) it makes explicit reference to the links with bolts in larger holes as in our case. I know that using 1.4 are more conservative, but I was interested in understanding whether I am wrong to interpret the norm or if there is any reason that I miss for which I should use instead the coefficient of 1.25 as reported in 6.5.8.1(3) for the ultimate limit state.

Thank you.

Hi.
♪

 

[Powermos]

checking the other accounts does not return to me the value of the precarious force or replacing the numerical values in the formulas I obtain:

0.7*84.3*800 = 47208 n

while I read 42208 n.

Did I miss something?

Hi.
♪

 

[Powermos]

I have some questions/questions about the calculation of fb,rd refractive resistance.

on the standard unv 1993-1 page 155 report [6.12] I do not understand the meaning of certain parameters in particular:

- was indicated the resistance to fracture for traction. In this case I have two different materials plates, one is in fe360 while the hinge bracket is in aluminum (I will return to the manufacturer the material and various characteristics). at this point the value of resistance to breakage, perhaps simplistic, but I think cautionary, should be given by the minor between the two materials used, is it correct? do you have any suggestions about this?
- d: does it represent the diameter of the hole or of the area suspended by the nut?
- t: It should be the total thickness of the union, in the case of two overlapping plates the sum of the thicknesses of the individual plates, right?

I apologize for asking questions, but being a novice of art, I prefer to ask a few questions, perhaps for granted rather than having doubts.

thanks for the attention.

greetings
♪