Bud's Clueless Remodel and Projects Thread

UPGRADING THE SHOWER HARDWARE

Got the shower hardware finalized today.

The guy who was in here last did some terrible work on the tile/grout/tile shelving, but that's a problem for another day...

Originally there was a basic low-flow shower head I replaced with a fancy new fixed/wand hybrid, but the wand wasn't magnetic and it bugged me enough to change things again. I've also gone back and forth on shower arm length/height and believe I've settled on the final setup now (maybe...)

Here is the most recent setup I'd used before today (Notice the cutout for the shower arm up top? Yuck):




For the final look I figured matte black would work better with the tile than chrome, or the brushed nickel I was in the process of switching over to. The flanges that come with shower arms are all about 2 1/2", which is too small to cover the messed up notching done to allow the shower arm to pass through; So I picked up a 3" flange to help cover most of the previous tenant's sins.

Here's a comparison shot of the various flanges; The silver is a stock flange size, the smaller black flange came with the black shower arm, and the larger black flange is the 3" model.




Here's a comparison shot of the various shower arms I've bought/installed. Hopefully this black 6" S-shape is the final configuration -- but you never know...




Here's a shot of everything installed:





Parts list:

• S-Shaped Shower Arm w/ Flange High Arc (Black 6")
  • Shower Arm Set Stainless Steel S-Shaped Shower Arm Concealed Shower Arm with Flange High Arc Bathroom Showering System Black 6"
    https://www.amazon.com/gp/product/B07SC8G7Z3
• Corla Modern Shower Arm Flange 3.5” XL (Black)
  • $18 Corla Modern Shower Arm Flange 3.5" Extra Large | Universal Escutcheon Replacement Cover Plate Round (Matte Black)
    https://www.amazon.com/gp/product/B08PV8MNB4
• Moen T2901BL Gibson Posi-Temp Pressure-Balancing Modern Valve Trim (Black)
  • $85 Moen Gibson Matte Black Posi-Temp Pressure-Balancing Modern Valve Handle Only Trim, Valve Required, T2901BL
    https://www.amazon.com/gp/product/B07N1H9DF3
• Delta Faucet HydroRain 2-in-1 Shower Head with 5-Spray In2ition (Black)
  • $288 Delta Faucet HydroRain H2Okinetic Dual Shower Head with Handheld 5-Spray, Hose, Detachable Shower Head, 2.5 GPM Flow Rate, Matte Black 58680-BL25
    https://www.amazon.com/gp/product/B07ZDPTZ95

Seems like everything'll be nice except the handheld unit is just about the right height to blast me in the face, but it's necessary to get the "rain" shower head high enough to suit my needs.

Edit: Can confirm that shower head is worth the money -- it's the best I've ever owned.

 

This looks like it will be a sweet apartment, does it have good views of the city? plenty of natural light with those large industrial windows.
Is the building owner going to offer you some free month's worth of rent for all your work on the place or is the rent that cheap that it's still worth it anyways?

 

It's got a pretty legit view of the city, which is unfortunately partially obstructed by a skylight on the floor below. Being just one floor higher would be 10x better, but regardless this place is still pretty sick and I'm happy overall.



If the owner wanted to reduce/waive some rent for work done that'd be great, but I've got the feeling he's kind of a cheapskate and would go the route of "nobody made you upgrade my dumpy unit..." -- which is true, I guess.

No love lost. This is more a quality of life thing where I don't mind too terribly much to eat the cost of some half-assed upgrades since I plan to be here for a pretty good stint.

I will be upgrading the stove/refrigerator in the future and would take those when I leave. Same goes for the bidet toilet I want to install -- that'd come along with me too since that's kind of a big deal and pretty expensive to replace.

 

REFURBISHING THE HIGHWAY SIGN (Part 2)

Got the freeway sign programmed and slapped together really quick for a party this weekend.

Put together using an Arduino, my very questionable C++, some wiring, and a Rubbermaid container as a temporary housing.

I'll post actual build details in a week or two when the supplies arrive and it's rebuilt properly. Until then, enjoy some contextless photos!

 

3D MODELING THE APARTMENT (Part 1)

Finally got around to taking a full set of measurements this weekend so I could get the apartment modeled in Sketchup. This'll come in real handy when planning the new kitchen cabinet layout, and figuring out how much shelving can be stuffed into the storage room.

Everything is pretty accurate, give or take 1/2-1" in a couple places; but that's just fine for rough spatial planning.

 

3D MODELING THE APARTMENT (Part 2)

Finally got around to adding furniture to the 3D model of the apartment/workshop.

The apartment furnishings are about 95% done, just missing a couple small pieces. Workshop is also almost finished save for the workbenches. Storage room is actually the first room I modeled to help lay it out and get a materials list for the shelving units -- it worked out excellent, down to the inch! The last major effort is the craft room/art studio that has lots of shelving needing modeled -- it's not difficult, just tedious.

Since planning for the storage room went so well, I want to also use this model to plan out workbench sizing/placement in the workshop, as well as cabinet/countertop layout for the kitchen remodel.

(Note: ceiling beams have been hidden to improve visibility into the model.)

Some living room shots:





The current kitchen area:



The storage room:



I'll post some final renders once the entire workshop + studio is fully fleshed out, along with some rough renders of future plans for the kitchen and workshop.

 

REFURBISHING THE DAIRY QUEEN SIGN (Part 1)

Can't seem to stop taking on new projects before finishing the existing stuff, oh well. This should be one that can get knocked out in a single afternoon.

1. Clean
2. Rewire with color changing/dimmable LEDs + an externally accessible power switch
3. Hang

Also made some progress on the highway chevron sign and got some parts ordered from TrafficLightWizard.com and TrafficLights.com -- those should be here next week. Still trying to find one more matching terminal for one of the light housings, but if that takes much longer I may just make a mold of one of the existing terminals to duplicate it...

 

A BIT OF HISTORY ON THE BUILDING

Spoke with a neighbor a while back who filled me in on the buildings past.

It used to be the Addressograph Multigraph building.

https://case.edu/ech/articles/a/am-international-inc

https://en.m.wikipedia.org/wiki/Addressograph

https://www.nytimes.com/1982/04/15/business/am-s-brightest-years-now-dim-memories.html

https://www.madeinchicagomuseum.com/single-post/addressograph-co/

 

As a current resident of PDX, you're really making me want to check out Clevland. Maybe you don't want to share but I'm curious what your place costs there. As you know, that's near impossible to find in PDX.

 

In PDX I got what felt like a great deal (but in a trash part of town) -- a whole ~700sqft house (+basement) near SE Cesar Chavez and Powell for $1,400/mo. It didn't feel like for that money you could really do any better without having roommates, or settling for apartment living.

In CLE the rent is about $1k/mo for a 4,000sqft loft just over 30 blocks east of Downtown in a district known as Asia Town.

Rent and overall cost of living in CLE beats the pants off PDX, but the nature out here does not hold a candle to the PNW.

 

RESTOMODDING A SCOREBOARD INTO AN ALARM CLOCK (Part 1)


Back in September I had an opportunity to buy an old scoreboard on Facebook marketplace. Not being one to look a gift horse in the mouth, of course I purchased it. At the time there was no real plan for its future beyond adorning the workshop wall.




WELL HUNG

First things first, this thing needs hung on the wall. This was accomplished using a series of pulleys and a Harbor Freight boat winch; That way it could be easily raised and lowered by one person so I could work on the restomod portion without needing anyone’s help moving it around.






FINDING PURPOSE IN LIFE

Once it was hung I promptly forgot about it for a couple of months – until inspiration struck. This is an old basketball scoreboard (Fair-Play BB-6620 to be exact) and has lots of displays for the various game stats, one of the obvious stats being a timer. That timer seems a perfect candidate for turning this bad boy into a big old clock. Heck, maybe one day I could even wire up the horn function and turn this into the world’s most obnoxious alarm clock!


Wiring it up as an alarm clock is cool and all, but what about the rest of the displays? Here’s the result of my brainstorming session below.



With a basic plan in place, we’re now in the “make it so” phase of this project. So… what do now?




GETTIN ER’ DONE

Lately I’ve been really into microcontrollers like Arduino and this project seems like a prime candidate for such a thing. Thus began the rabbit hole of research.

These boards are usually updated via a proprietary controller sold by the same manufacturer. The first problem with such a piece of hardware is it is meant to be used for scoring an actual sports game and not as a weather station/clock interface and are manually updated by a user. The second problem is they cost around $350 in addition to not being flexible enough for this project. All that to say: this project probably won’t be using a sanctioned controller to update the display.


Without an actual controller to update the display, there seems to be a couple options on how to proceed:

• Rewire the displays using multiplexed relays connected to a(n Arduino) microcontroller and cut out the scoreboard logic entirely
• “Emulate” a controller using a custom (Arduino) microcontroller and trick the scoreboard into thinking it is getting commands it is accustomed to

OPTION A - EMULATION

Emulating a controller seems like an awful lot of work based on initial research into how such a thing is accomplished. You see, commands are sent via ¼” jack into the scoreboard using serial communication and it is not very well known exactly what format the data is communicated in. There are a couple of GitHub projects that shed a little light onto the issue, but it seems like something that would be a royal pain in the ass to implement.

Here is a project showing an example of decoding the output from the exact controller used on this specific model of scoreboard: https://github.com/will62794/MP-69D-Scoreboard-Decoder

While I’m very grateful to will62794 for taking the time to share this, the documentation is a bit lacking within the code, so reverse-reverse-engineering this to not decode, but instead encode messages is just a bit messy.

There is another project showing an example of encoding messages available here: https://github.com/evm-sec/FairPlay/blob/master/wf-pico.c

Unfortunately this example is specific to baseball and is again very lacking in the documentation, which leaves a lot of inference to whoever chooses to make use of it. Now, you could cross-reference the decoder project along with this project in order to figure out the data format sent to the board and all that, but even after all that, there is still the issue of wiring this all up once it (possibly) works. will62794 alluded to the input being decoded passing through a TI AM26LS33 receiver chip before being passed to the Arduino, which implies that there is a similar wiring necessary to send messages to a scoreboard. Welp, that is beyond me, so this method of emulating a controller is dead in the water.


OPTION B - BRUTE FORCE

With emulation off the table, it seems like gutting this and rewiring all the displays to be controlled with relays is the most “practical” option. This is non-optimal since it will irreversibly alter the internals of the board from their original state to this hack-n-slash wiring mess – not that I plan to ever reverse this modification, but it just feels a bit wrong is all.

This board has eighteen digit displays in addition to four displays I will be referring to as “binary” as they are simply on or off (bonus/possession indicators.) Two of the eighteen digits are single segment displays for the number “1”, the remaining sixteen are eight segment displays for “0-9”.




Here’s a quick “back of the envelope” table of which relays are powered to create which digits.



This means that to run this board and control all digits there are (2*1)+(8*16) or 130 relays required. This is a setup that’s far from impossible, but will require quite a bit of hardware to accomplish. Without getting into the full shopping list, here are the major components required:

• (1x) I/O Expander – $40
• (10x) Relay Expander – $100
• (10x) 16 Channel 12V Relay Module – $150

That list is already up to nearly $300 in parts and that’s before the additional ~$250 in Molex wire adapters, ~500ft of hook-up wire, and wiring terminals with their various accoutrements. So this route is looking pretty expensive. If everything went without a hitch, it’d be nearly $900 all-in on the board and materials for rewire.

This option was dead to me once it came clear that not only would it be the most expensive, but wiring ten relays from 120V AC to drive arrays of 2.5W bulbs would require even more hardware to step the power down to a level that wouldn’t fry the bulbs and their boards.


OPTION C - MAKE PEACE WITH THE FACT I’M TAKING THE HARD ROUTE

So here I find myself contemplating emulation of a controller again and feeling pretty defeated about the whole ordeal. This lead to another round of intense googling, which netted a very promising thread on the Arduino forums where another user was attempting almost the exact same project on a scoreboard of his own; which ultimately ended up working exactly how he wanted. What luck!

Here’s a link to the post: https://forum.arduino.cc/t/reverse-engineering-basketball-score-board/1003225

Toward the bottom in post #43 on the thread, Edsjunk says:

To summarize, Ed’s suggesting cutting out all the noise of actually emulating a controller via serial communication, and simply directly manipulating values on the scoreboard’s chipset.

Welp, “shift registers” and “74HC595”s are far beyond me, but that’s why we have Google. What Ed’s basically saying is there is a well established concept of shift registers which are basic chips with a small amount of memory that can be used to store and send small bits of information. Thankfully there are a plethora of examples online for controlling a 74HC595 using Arduino. One such example can be found here: https://docs.arduino.cc/tutorials/communication/guide-to-shift-out

The main thing here is that these registers have three key pins used for interacting with them: data, latch and clock. All the tutorials relating to shift registers and Arduino make use of these pins. The diagram below shows each of those pins on the 74HC595 chip.



That’s all good and well, however notice how in Ed’s post he asked “Could I control these shift registers as I would a standard 74HC595 Shift register?” The key phrase here is “as I would”, because the scoreboard uses a shift register, but it is not the 74HC595. So which register does it use? Let’s take a look…



It seems this board uses the HEF4094BP chip –and I’m no rocket scientist, but– that seems like it’s probably not the same as a 74HC595. So now it’s time to google this new chip and find its pinout. Thankfully DigiKey has this covered with a datasheet for the chip: https://www.digikey.com/en/products/detail/nxp-usa-inc/HEF4094BP-652/1023605

Spoiler: AUTHOR'S NOTE

It seems the HEF4094BP chip has "strobe" as the #1 pin, unlike the 74HC595. After some reading, it looks like the actual "latch" pin may be the #1 "strobe" on this particular chip. This'll take a bit more digging to ensure whether/how the #15 "output enable" is used.

Now the only question left is this: how do you know which pin is #1 and which pin is #9? Well, take a look at the pinout diagrams and you will notice a circle in the upper left of each diagram. Thanks to EvilMadScientist, I now know that is what’s called a polarity mark. Pin #1 will always be located at the first pin traveling counterclockwise from that mark. So where is the polarity mark on the HEF4094BP? Great question. Let’s see what EvilMadScientist has to say about that: https://www.evilmadscientist.com/2010/basics-finding-pin-1/

Looking at the photo of the chip from the scoreboard, let’s find pin #1.



Now I’m armed with a possible solution thanks to Ed, the correct pinout for the shift registers used on the scoreboard thanks to DigiKey, and know which pins to target compliments of EvilMadScientist. The next step is to identify which chip is actually “the first shift register” Ed is suggesting connecting to in his post, and half-assedly connecting an Arduino to it in order to start throwing junk data at it in the hopes of manipulating the values displayed.


THAT’S ALL FOR NOW…

Thanks for sticking around so far. Sometime soon I’ll post an update, hopefully with good news about how things are progressing on the project.

 

Will digest later.

Wow though.

 

This is way beyond my abilities -- and also way cool!!

Awaiting the next update.

 

Girlfriend has been asking me to buy a bar cart for a couple weeks.

Found this on Facebook Marketplace and did a little road trip with the dog today to pick it up. Now the apartment has a "bar cart" -- it still needs some silicone mats or something so the bottles aren't clanking on bare metal but it's basically good to go.



Came out of a salvage yard garage, apparently they don't need it anymore. It has five bins but I'll only be using three for it's new life as a cart.

 

Yesterday the parts to restore the GameStop sign came in -- two transformers (one for each word), and a dimmer which'll help tone down the brightness since this is indoors now.

Unfortunately I was too excited working and took zero progress pics, which sucks because I'm pretty dn proud of the wiring job. The power is a normal 3 prong extension cable but it's plumbed through the wall so there's not an unsightly cable hanging down.

Also got the old TV wall mounted as a shop TV since I was in a doing mood.

 

Added this little lady to the family back in November. She was wandering in the cold, emaciated, dehydrated, mangey.

She's had quite a glow-up since then. Thanks to her overpriced food she's filling out and adding lean muscle, her coat is improving and she has the run of this place. It took a couple months but she's come into her own and loves to play almost as much as she likes to nap.

Meet Tofu:

 

Good people give lonely animals homes, made my day, thanks!

 

don't mind me, i'm just here for the dog pics!

 

RESTORING AN OLD FIRESTONE SIGN (Part 1)

A DIAMOND(?) IN THE ROUGH

Like many of the projects cluttering the workshop, this one started with me mindlessly browsing Facebook Marketplace looking for a new toy to sink money into. Collecting and restoring old signs is becoming somewhat of a hobby for me these days, and this sign was listed for a half decent price.



The pictures seemed to show channel letters in good condition, with missing or broken neon tubes, rusted and blown out neon transformers, and missing or shattered acrylic and trim capping.

Needless to say I packed up the dog and we took a Sunday afternoon road trip to western PA to grab this bad boy.


WHERE TO START?

If you've ever handled these channel letters that are lit by neon, you're well aware the transformers add quite a bit of heft to an otherwise feather light assembly. So to kick things off, it seemed like a good idea to clean up the housings by removing the old neon tubes, the transformers, and the glass posts used to hold the tubes in place.



Here's one of the letters with the most intact insides showing what needs removed. Somehow it was my lucky day and these absolutely rusted out transformers all came out without a single broken bolt and a minimal amount of PB Blaster.



The little glass posts securing the neon tubing in place were secured using metal L brackets that were riveted to the housing. In what is yet another small miracle on this project, those rivets were aluminum and were able to be drilled out without the slightest bit of hassle.



The next step was to take the now empty carcasses over to the slop sink and give them a good scrubbing with some hot water and Dawn dish soap to get the years of crap off the insides and outsides.


NOW WE'RE GETTING SOMEWHERE



With the prep work out of the way, now the actual project work can begin. Since this will end up mounted on a wall like the GameStop sign, wiring each letter individually is a bit impractical and would be difficult to accomplish in an aesthetically pleasing way.

The GameStop sign has a nifty backing behind each section so rather than mounting eight individual letters, you're mounting two larger groups of lettering. The backing also provides a convenient place to tuck away any unsightly wiring.

These letters obviously didn't come with a backing plate since they were originally mounted directly to a tire center; So I will need to find a way to fabricate one.

TEDIUM

To make a backing plate that looks anywhere near acceptable, it'll need to be cut on waterjet or by laser, and those will need a CAD file so they know what to cut. Free handing something like this is not an option.

My first hunch was to go to the Firestone website, snag a PDF with their logo in it --since PDFs are usually vectors and you can scale the logo in them infinitely without losing quality; rip the logo using Adobe Illustrator, and Bob's yer uncle.

Unfortunately I quickly found out the font used by Firestone today is similar to but still very different from the font they used back in the day. Best I can tell, this sign is from about the 1950s based on the transformers, and some old Firestone advertising materials I've found with a very close (but still different!) font. So the PDF route was dead.

Ripping logos off older advertising materials was also a no-go since none were a perfect match for this sign. So then how do I take the sign from meatspace into the digital space with any degree of accuracy?

Simply snapping photos of the channel letters with a phone and tracing them is a no-go since there will be a large amount of distortion due to perspective being a thing and there being no way to take a photo from perfectly directly above each and every letter...

Then I had a light bulb moment. Why not take a box of known dimension, cut it to the same depth as the channel letters, place each letter inside that box, snap a photo, take each photo into Photoshop to remove the distortion, and then trace the result? It's crazy enough it might just work!

Here is an example showing what a photo looks like originally, and after removing the perspective distortion.



It's not a huge difference, but without the step of removing distortion the backing would be completely misshapen, and I would have no idea where to place the through holes from the back of each letter.

Why cut the box to the same depth of the letters? Because if they were not the same height the letters would still be distorted, even if I modified the photos to square up the containing box -- just look at the difference of the workbench between the original and un-distorted photo to see what a difference only a few inches of depth makes.

The final result of un-skewing each letter is below.





I was able to take these images into Illustrator in order to trace the lettering and plot out where the through holes should be. Once the letters were traced, it was relatively simple to add a border around them which would be the final shape of the backing.

Here is an example of a traced letter with the through holes also plotted. The through holes are larger to give a margin for error on all this image editing and tracing work.



And here is the final result, showing what the final tracing looks like with lettering superimposed, as well as without which is what the final backing plate will look like.






TESTING THE THEORY
Ok so eyeballing it everything seems to look pretty accurate. But if the laser cutting is going to cost $200-$400, it'd be nice to know for sure that everything was relatively accurate and lined up before blowing a wad of cash on something that might not be usable for the project.

So I ran off to FedEx office where they have a big fancy printer that can print 3'x100'+ and got this thing printed out at full scale to see for myself if it worked.





Well I'll be damned! Got it bang-on the first time. There was one through hole with a slight clearance issue but everything else lines up perfectly. After seeing this at full size, I decided to trim back the outline by about 1/3 so it wasn't so aggressively large.


LEARNING NEW THINGS

Almost every project requires me to learn some new skill before it's over and this sign is no different. Metal fab shops want DXF files to feed their machines and brother I've got no idea what those are. Luckily Google does... they're apparently some sort of CAD file, and Fusion360 can export them.

So I popped over and grabbed a copy from the Fusion360 website and opened that bad boy up. To date my only experience in 3D modeling is with Google Sketchup which is a relatively basic CAD tool when compared to Fusion360, SolidWorks or any of the big boy softwares.

Thankfully this project is really simple and I'm just importing the SVG from Adobe Illustrator, extruding it, and then converting it to a "sketch" -- whatever that is. The hardest part was ensuring the scale was correct in the model so that these fab shops would be able to provide an accurate quote.



There she is in all her glory, a fancy CAD file that can be sent to fab shops! I've already got a quote and am visiting a shop tomorrow to see how they actually create this stuff and pay for the service. Exciting stuff.


NEXT STEP(S)

Once the backing is fabricated, the last bit of this project is that pesky little "restoration" step. I've already purchased tools and materials for that part, but this post is already godawful long and that part can wait for another day.

 

Anxiously awaiting part 2 - this is way cool!