Archive for the ‘Uncategorized’ Category

Bunnings site rip

Some time ago I fool­ishly volun­teered to per­form a site rip of https://​www​.bun​nings​.com​.au/ for the local SES group I am a mem­ber of. This was to allow our account­ant mem­ber to more accur­ately assign a value to our assets. I under­stand this is an import­ant thing for an accountant.

I have done a num­ber of site rips in the past, the Bun­nings site is prob­ably the most pain­ful so far. The product pages are very com­plex for what they are.

Each Bun­nings product page is roughly 300k. I extrac­ted 1.1k of con­tent from each page. So 99.63% of it basic­ally use­less, or an effi­ciency rate of 0.4%. The vast major­ity of the space is taken up by the nes­ted menu at the top, the ads near the bot­tom take a bit and then there is a fairly extens­ive site map across the bot­tom. At least the CSS is in an external file, well, four of them.

There is a mobile web­site which is a bit slim­mer. I think the page served is triggered by browser fin­ger­print­ing and cook­ies. I didn’t dis­cover it until too late though.

There are also two dif­fer­ent HTML struc­tures used for product pages, they look sim­ilar but have dif­fer­ent tags with dif­fer­ent classes.

And a fun trick, these two links go to the same page:
https://​www​.bun​nings​.com​.au/​r​o​m​a​k​-​m​6​-​h​i​g​h​-​t​e​n​s​i​l​e​-​c​o​u​r​s​e​-​h​e​x​-​n​u​t​-​1​0​-​p​a​c​k​_​p​1​1​0​0​797
https://​www​.bun​nings​.com​.au/​n​o​b​o​d​y​-​n​i​b​b​l​e​s​-​n​u​t​s​-​l​i​k​e​-​n​o​d​d​y​_​p​1​1​0​0​797

That trick gets less awe­some when you real­ise that they actu­ally do this and link to the same product with dif­fer­ent urls, 626 times.

In case any­one else is feel­ing fool­ish enough to try this them­selves, and brave enough to look at my code, the end res­ult of my tri­als and tribu­la­tions is on git­hub. All the mis­takes have of course been purged from the his­tory so it looks like I just bril­liantly did it in one go.

https://​git​hub​.com/​l​o​d​/​b​u​n​n​i​n​g​s​-​s​i​t​e​rip

Microwave project managing – week 2

Man­u­fac­tur­ing

The visit to AMAPRO on Fri­day was a sig­ni­fic­ant suc­cess, I feel con­fid­ent enough about the pro­cess that I didn’t fol­low up my other tur­ret punch­ing options.

The com­pany is run by a man who strongly reminds me of someone I have worked for recently, and to a lesser degree an earlier boss. This sim­ul­tan­eously reas­sures me that he will do a good qual­ity job and also ter­ri­fies me because I am where he extracts his money.

The fact­ory floor tour was very impress­ive, an expans­ive space but very clean, no clut­ter. I was most inter­ested in their three Amada tur­ret punch machines, recent mod­els which seemed well main­tained, one was also fit­ted with an auto­mated feed allow­ing it to run unsu­per­vised overnight. They also had a num­ber of press brake machines, some auto­mated weld­ers and a range of other stand­ard tool­ing toys.

The com­pany has two engin­eers that do design work, includ­ing devel­op­ing their own AMAPRO products. I will prob­ably use them to assist in shift­ing from my sketches to a man­u­fac­tur­able product.

It was rather dif­fi­cult to dis­cuss price estim­ates, partly because it was very early in the design pro­cess, I also got the feel­ing that any final quote will vary greatly based on the quant­ity. They also offered that pri­cing for pre­lim­in­ary design and tool­ing work could be based on the even­tual final order, some­thing I am wary of because I intend to get at least one other quote for the final job.

The costs looks at being above $150 but haven’t taken into account the front face. For the ini­tial cal­cu­la­tions I am going to assume the front will also be metal, to avoid the plastic tool­ing costs, and work with a chassis price of $200. This is a sig­ni­fic­ant per­cent­age of my tar­get cost, higher than I was hop­ing for, but the cost struc­ture does have the advant­age of being fairly lin­ear based on quant­ity. I also think I will be able to offer the choice of cheaper metals with enamel and powder coat­ing or full stain­less as the tool­ing is identical. There are some other options which may be able to get the even­tual cost down but they come with con­sid­er­ably more risk.

The mech­an­ical por­tion of the feas­ib­il­ity stage is now com­plete. When I get around to pro­to­typ­ing I will prob­ably build the first pro­to­type myself in the gar­age, then draw up a formal design and do one or two rounds of small run man­u­fac­tured prototypes.

Thermal Ima­ging

I also spent some time invest­ig­at­ing thermal ima­ging sys­tems. I have decided on two devices to buy and invest­ig­ate fur­ther, both are very different.

The first is the Seek Thermal cam­era. This is a phone attach­ment which retails for $250 US, the latest after a series of Kick­star­ted products and an entry from FLIR. The com­pany is inter­est­ing, the founders have a solid his­tory in the industry, their prior com­pany was pur­chased by FLIR and they recently got a healthy check after FLIR tried to attack them through the courts. The cam­era seems amaz­ing, sig­ni­fic­antly bet­ter res­ol­u­tion than any­thing else on the mar­ket under $1000 and has a thermal range up to 350 ° C, many oth­ers cap out at around 100 ° C. The sub­stan­tial down­side is that Thermal makes their own sensor pack­age and don’t sell it as a com­pon­ent, I am going to have to talk them in to chan­ging this. It is also rather expensive.

The other is the Panasonic Grid-Eye 8×8, a com­pon­ent which provides a 64 pixel array of tem­per­at­ure. The main advant­age is that it is price com­pet­it­ive, $39 USD in single unit quant­it­ies, the eval­u­ation board is only $75 USD. The dis­ad­vant­ages are sub­stan­tial, the res­ol­u­tion is poor and the tem­per­at­ure range caps out at 100 ° C. How­ever while the pic­ture won’t be nearly as good to look at com­pared to the Seek Thermal I think the res­ol­u­tion will be enough for the func­tion­al­ity I envis­age and tar­get tem­per­at­ures tend to be under 100 ° C.

Worth men­tion­ing is the MLX90620, my third option. This was used by the IR-Blue pro­ject, one of the early Kick­star­ted infrared phone products. The sensor pro­duces a 16×4 pixel array with a tem­per­at­ure range up to 300 ° C. How­ever at $87 USD per unit it sits in an uncom­fort­able space between the two other products, bet­ter than the Grid-Eye but nowhere near as good as the Seek Thermal. I may revisit it if the other two don’t work out.

Dis­play

I was hop­ing to find one or two options for the dis­play this week but didn’t get around to it.

View full screen

Microwave project managing – week 1

Mech­an­ic­als

So the last week has been spent dom­in­ated by get­ting my head around mechanicals.

I had fant­astic vis­its to three metal stamp­ing man­u­fac­tur­ers, Diver, IXL and Com­pon­ent Indus­tries. All three were help­ful well bey­ond expect­a­tions, espe­cially since they real­ised very quickly that I wouldn’t be able to use them.

I under­stood that metal stamp­ing had upfront costs which were off­set by the reduced unit costs. What I didn’t under­stand was the scale that they work at. Stamp­ing is good once you want 10,000 of some­thing and really sweet once you add another zero to that num­ber. My pre­lim­in­ary work­ing num­ber is 500, so I have to use dif­fer­ent techniques.

The lower cost method is a com­bin­a­tion of tur­ret punch­ing, laser cut­ting and fold­ing. I have actu­ally played with sim­ilar tech­niques in a metal sculp­ture course I did a few months ago, I just didn’t think they would come into play for this kind of product. I have vis­ited a laser cut­ter, Bey­ond Laser who prom­ised that if I could draw it, he could cut it. I also have a visit lined up with AMAPRO and Cono­matic for Friday.

It looks like what I am after is pos­sible. I should have a much bet­ter idea on Fri­day but it seems like I am look­ing at $100-$150 for the mech­an­ic­als. This will involve more labour than the stamp­ing tech­nique and it may be worth look­ing over­seas to reduce the costs.

There are also some sig­ni­fic­ant unknowns with the mech­an­ical design that impact the cost, such as what hap­pens if you put sharp corners in the microwave chamber.

Invert­ers

The other main task was look­ing in to using inverter tech­no­logy. This was less fun. Basic­ally the tech­nique is tied up with pat­ents, primar­ily held by Panasonic. The non-Panasonic inverter microwaves seem to be either licenced or rebadged Panasonics.

Being able to buy the Panasonic guts would actu­ally be ideal for me and seems like fairly stand­ard prac­tice in the industry. Fig­ur­ing out who to approach is not easy though, I have sent one email off and will see if it leads to a way in.

View full screen

Return top