85 replies to this topic

[Johnny D]

You guys are going a little steep.

 

I bought one of these loadcells for a test: https://www.amazon.c...1_O0gozb0RDP7V8

My weigh plates are prototyped in ply wood for now. Basically I'm imagining 1/4 steel plate approximately 12"x12" or similar. 1/4 carbon steel plate is about $10 a square foot.

 

So I get this so far...

4x 1k loadcells = $200 - https://www.amazon.c...1_O0gozb0RDP7V8

1x RPi3 = $36 - https://www.amazon.c...1_L9gozbNQBY5F4 (could be swapped for RPi Zero W for 10 dollars less)

4x Loadcell ADCs = $8 - https://www.amazon.c...1_z.gozbGZ390B2

1x PCB = ~$12 - Prototype boards in low quantity can be ordered in minimum of 3 units @ approximately $37 dollars (Optional cost, you could wire the electronics with no board)

8x steel plates = $85 - any local fab shop should have something like this; no leveling mechanism

Misc hardware = $20 - bolts, etc

 

Each set of plates requires 3 holes to be drilled for mounting load cells that can be done yourself.

 

Total cost is $361.

 

That is bare minimum, but we could actually work on this platform and extend functionality etc. I don't know if the other scale companies even have a phone app wireless interface or ability to store measurements etc, but they definitely are not on their entry level products.

 

If anyone has an idea of how to get the cost of the plates lower that would be huge since electronics are already dirt cheap and load cells probably won't be easy to get cheaper.

 

Cool, I'm from Missouri (show me)

 

J~

[Steve Scheifler]

You'll need to draw me a picture of the steel & load cell sandwich.

12x12 is kart size. Need 15x15 or it is too much hassle getting the car close to centered on all four, and turn plates are 15".

Like I said, a fun project but more work and $ than it's probably worth to get something accurate enough for cross-weight adjustments.

Speaking of which, water levels are OK but don't neglect getting each scale leveled individually.

[OrangeCrush86]

I'll share when I have a prototype plate finished. Getting 0.1% accuracy of the load cell full scale should not be that difficult.

[Andy Mitchell]

How much will these things weigh? I'm not sure that 1/4 plate would be rigid enough in the first place (there are going to be big twisting moments at the mount points with 500 - 1000 lbs load), but even so, a chunk of 15" x 15" x 1/2" steel sounds like it would be a bit hefty to move around easily. Four of them together might not be that easily portable.

 

Not trying to be negative, just thinking out loud. Interesting project.  

[Andy Mitchell]

Whoops, double post. Excuse my clumsiness, please.

 

The second post was identical to the first but added a comment that the rigidity I was concerned about was the torsional stiffness at the load cell mount point. A load on the scale produces a twisting moment there equal to the load being weighed x the length of the load cell. And even a single degree of twist at that mount will result in a significant downward deflection of the scale pan. Anyone know how much deflection in a race car scale set is tolerable, or how much the commercial versions move under a typical load?

[Brandon]

Dirk's link to McMaster's AL sheet plate page (https://www.mcmaster...sheets/=17yuh4g) is illustrative of what our options would be for aluminum instead of steel.

 

Uncertain what type of metal we'd want (they have all manner of options it seems; corrosion resistant, hardened, multipurpose) but they appear to have the sizes & thickness we'd need.

[Steve Scheifler]

McMasters is a great 1-stop source, but if this is all about value it may not be the best place to buy

[Brandon]

McMasters is a great 1-stop source, but if this is all about value it may not be the best place to buy

 

Don't disagree however that would give everyone a gauge of the top-end pricing for the self-sourced materials.

What are the thickness of the plates from Lonacre or any other manufacturers?

[Bench Racer]

What are the thickness of the plates from Lonacre or any other manufacturers?

If you ment to say plates from LongAcre. The scale platform is not a plate. It is a honeycomb aluminum casting. Webb thickness is 1/8 inch and each overall platform thickness is 1 1/8 inch. See inside of platform honeycomb configuration.  

 

https://www.bing.com...x=29&ajaxhist=0

[Jim Drago]

Also made my first liquid level. Bought my second 8-10 years ago for $27.00, today it's $29.99.

http://www.speedwaym...-Hose,2702.html

After much ball busting.. I clicked your link and realized it wasn't Bennetts home made pos..  I ordered one and will check and use at the Sprints and report back  If its more accurate.. I will eat crow and give you your due  

[BNaumann]

After much ball busting.. I clicked your link and realized it wasn't Bennetts home made pos.. I ordered one and will check and use at the Sprints and report back If its more accurate.. I will eat crow and give you your due

It's perfectly accurate because yay science but a bit cumbersome to use. I slip a 2.5lb dumbbell weight over each one to keep the memory of the not-so-flexible hose from dragging the cylinders across the floor. I fill it with windshield washer fluid because it is dyed enough to see but still easy cleanup. In the long run I used it to check calibration of a $50 laser level. Turns out the laser level is dead nuts in a straight line but a bit far off left to right. Still works good enough even after I crushed it with the car.

[Bench Racer]

After much ball busting.. I clicked your link and realized it wasn't Bennetts home made pos..  I ordered one and will check and use at the Sprints and report back  If its more accurate.. I will eat crow and give you your due

Without comment, I was going to bring to Sprints. Altho sometimes things seem a bit tight under your tent, plus food shortage. Busting balls with Bennett is another story, I don't go there.

 

It's perfectly accurate because yay science but a bit cumbersome to use. I slip a 2.5lb dumbbell weight over each one to keep the memory of the not-so-flexible hose from dragging the cylinders across the floor. I fill it with windshield washer fluid because it is dyed enough to see but still easy cleanup. In the long run I used it to check calibration of a $50 laser level. Turns out the laser level is dead nuts in a straight line but a bit far off left to right. Still works good enough even after I crushed it with the car.

You know your not allowed to get in on this water level ball busting between Jim and I which has been going on for several years. 

[Bench Racer]

Guys, with reference to the stiff hose (& trust me I like a stiff hose) Speedway Motors sends with the liquid level, I went to a home fish tank/stuff supply store and bought some really soft flexi hose. Zero issues with twisting/kinking/silly stuff with hose.

[Alberto]

You guys are going a little steep.

 

I bought one of these loadcells for a test: https://www.amazon.c...1_O0gozb0RDP7V8

My weigh plates are prototyped in ply wood for now. Basically I'm imagining 1/4 steel plate approximately 12"x12" or similar. 1/4 carbon steel plate is about $10 a square foot.

 

So I get this so far...

4x 1k loadcells = $200 - https://www.amazon.c...1_O0gozb0RDP7V8

1x RPi3 = $36 - https://www.amazon.c...1_L9gozbNQBY5F4 (could be swapped for RPi Zero W for 10 dollars less)

4x Loadcell ADCs = $8 - https://www.amazon.c...1_z.gozbGZ390B2

1x PCB = ~$12 - Prototype boards in low quantity can be ordered in minimum of 3 units @ approximately $37 dollars (Optional cost, you could wire the electronics with no board)

8x steel plates = $85 - any local fab shop should have something like this; no leveling mechanism

Misc hardware = $20 - bolts, etc

 

Each set of plates requires 3 holes to be drilled for mounting load cells that can be done yourself.

 

Total cost is $361.

 

That is bare minimum, but we could actually work on this platform and extend functionality etc. I don't know if the other scale companies even have a phone app wireless interface or ability to store measurements etc, but they definitely are not on their entry level products. Honestly with a little more work and another $100 in electronics we could even have wireless battery powered pads.

 

If anyone has an idea of how to get the cost of the plates lower that would be huge since electronics are already dirt cheap and load cells probably won't be easy to get cheaper.

 

In the end this is just an experiment for me. It's entirely possible that we can't beat the manufacturers that can keep costs low with mass production.

 

A local guy built a set of awesome 4.1xx hub stands with support for toe measurements and camber.  He's about to go into production at about $6xx a set of 4.  Combine with your scales, this may be a nice grassroots option.

[Steve Scheifler]

Pictures?
I've seen other attempts at this but I don't understand how it can ever get you to a final setup, inluding cross-weight, given the interaction between the various adjustments. Perhaps that's why they never catch on?

[dstevens]

It's a good idea.  I've been involved in the open hardware community for about 7 years now, the last 5 owning an open hardware based fabrication company.  For you guys looking just to get something on the cheap this isn't the project for you.  Designing and building open hardware is more than just looking to get something on the cheap.  It's a process and until it gets past the first several iterations and becomes more mature it's not for end users but early adopters.   For guys like Benchy and Steve, this isn't for them.  These sorts of projects are for people that want to tinker and experiment and do their own thing.

 

The mechanical implementation of the scale pads are likely to be the big cost driver.  The load cells aren't bad, you can get them for $20-30 direct from Taobao or Aliexpress.  If you were in Shenzhen you could probably get the cells, connectors, cables, parts for the boards and Pi for under US$100 in the market at HQB. I forked it and took a look at it.  Not a lot of people in the open hardware space use Fritzing for the boards.  The gerbers in the repo aren't saved as gerbers and fab houses want files in a gerber format and not a pdf of the gerber.  I can put them into Eagle and get some boards made.  The boards are about half a buck in a 10 qty plus shipping from Shenzhen.  

[OrangeCrush86]

It's a good idea.  I've been involved in the open hardware community for about 7 years now, the last 5 owning an open hardware based fabrication company.  For you guys looking just to get something on the cheap this isn't the project for you.  Designing and building open hardware is more than just looking to get something on the cheap.  It's a process and until it gets past the first several iterations and becomes more mature it's not for end users but early adopters.   For guys like Benchy and Steve, this isn't for them.  These sorts of projects are for people that want to tinker and experiment and do their own thing.

 

The mechanical implementation of the scale pads are likely to be the big cost driver.  The load cells aren't bad, you can get them for $20-30 direct from Taobao or Aliexpress.  If you were in Shenzhen you could probably get the cells, connectors, cables, parts for the boards and Pi for under US$100 in the market at HQB. I forked it and took a look at it.  Not a lot of people in the open hardware space use Fritzing for the boards.  The gerbers in the repo aren't saved as gerbers and fab houses want files in a gerber format and not a pdf of the gerber.  I can put them into Eagle and get some boards made.  The boards are about half a buck in a 10 qty plus shipping from Shenzhen.  

 

Hello and thanks for the interest. I know Fritzing isn't the best option. I only did it because I already started there for a graphical view and i didn't have time to learn Eagle (this is my first PCB design). If you can make files in Eagle for the board design that would be a big help. If you have a source for some cheap boards I would like to work with you. I would like to manufacture a few after I do small update to the board so HX711 channel B can be used.

[dstevens]

I initiated a pull request that has rs format gerbers generated from Fritzing for your repo.  I ran them through the gerber viewer at Seeed and they passed.  I almost bought a batch last night but decided to wait to see if that rev was rtm.  If you put it in another branch when you finish I'll merge it to my repo and generate some gerbers and place an order.  Cost (which I'll gladly pick up) is about $5 for the parts (we get a nominal 10 parts per order) and $20 for DHL from Seeed.  Dangerous Prototypes is about $17 including shipping but they'll take a week or two longer via China post.

 

I was going to convert to Eagle last night but I don't have a footprint for the NX711 carriers.  I'll have to make one.  It was my first time using Fritzing.  Pretty cool, much more maker friendly than Eagle.  Sparkfun has a Fritzing library. 

 

I haven't put those python scripts on a Pi yet.  I've only got 1 Pi3, a few Pi2 and those are being used in production but I've got some B+ laying around (and a Pi TFT).  That should be enough to demonstrate the concept.  Once I get that up and working I'd like to try it with a Pi ZeroW.

 

Right now I'm slapping a new coat of paint and cleaning up the SM I built and never raced so I can try and sell it as a Heritage class or entry level SM.  I'm also starting a build with my wife for a class 7100 truck for the 2019 Mint 400 so we can run it as more or less tourists.    I've still got some time I can work on this every now and again.  Python isn't my strong point but on the electronics side I've got enough game to help get something like this prototyped, packaged and ready as a maker project.

 

I'd like to see if someone out there in the racing sphere had an old or broken 15" scale pad that we could use as a design reference.    A donation to the project would be good but I'd be willing to pay a small amount and pay for shipping.  Steel plates are certainly workable but something lighter would be better.  I think getting the usability, utility and cost of the pads in line is going to be the most pressing long term part of the project.  Even dims and pics from an existing pad would be useful.

 

When I was using AL for the 3D printer frames we built I was getting .250" 6061-T6 for about $260 a 4' x 10' sheet.  It's been a year and a half or so since we switched to another material and AL has gone up in that time.  We had a 16" footprint for the cut parts and we got 35 units from a sheet  IIRC (I'll have to pull that nesting file up and see).  For base and top pad material it's about $65 plus the cutting, waterjet, EDM or a larger laser.  For 3/16" steel it's about $20 a set plus cutting.  And a bunch heavier but cutting is way cheaper in bulk.

[Johnny D]

So orange said and I'm waiting....

"8x steel plates = $85 - any local fab shop should have something like this; no leveling mechanism"

 

Stevens... So you really only need 4'x4'x1/4 Al and you get 9 pcs 16"x16" and if you look for the cheapest, at McMaster...

looks like 5086 grade Al at $293. no tax, shipping or cutting.

https://www.mcmaster...sheets/=183164n

 

J~

[dstevens]

McMaster is great for prototyping, not so much for production.  I get a McMaster order every week or so but I wouldn't get small batch metal from them.  You can get a full sheet at your local metal supply for about that price.   Shaprio Supply in St Louis is a better choice for small batch.  https://www.shapiros...t-plate?cat=472

 

Steel is pretty cheap.  When I was going to the Techshop locations up your way I used Campbell in San Jose.  The pricing is more than it is in Vegas (the delivery was stupid expensive) but the selection is good.  Shapiro is still a better bet small batch.  https://www.shapiros...d-plate?cat=493  The last 4'x10' 3/16" steel plate I got at Curtis locally was $120 or so delivered.  I can get pretty much anything I need at Curtis except 4130 and DOM.  Unless you are cutting with a band saw, hand saw, sawsall machine cutting steel is much cheaper.  To waterjet that square may be $10-12 in cycle time per piece.  You can plasma cut that for a buck or two.  Of course the shop is likely to have $100 or more minimum.  If you are a Techshop member the plasma table in Redwood City is part of your membership and the waterjets at SOMA and San Jose are about $3 a minute.  A sawsall and a vise would make pretty quick work out of a project like this.  One thing to consider on a maker project is to use tools that others have access and the skills to use.

 

I haven't looked in my Shigley's or Machinery's Handbook at what grade AL to spec just yet but I'd go to the side of 6061-T6.  5086 might be too soft though I'd have to lean on a better materials guy than I.   Once I get some sort of base design it can be simulated in a low buck FEA in Fusion 360 although a better test would be to make one and just drive a car on top of it.  I'm thinking something like two plates separated by some steel tubing in each corner, HREW should work.  You could do a leveling mechanism but you'll still need ramps and roll offs to do it right so I don't know if a built in leveler makes sense.  What's cool about the open hardware scene is you can take the design and hack something like the leveler into it.