Jump to content

jovial_cynic

Senior Member
  • Content Count

    761
  • Joined

  • Last visited

Community Reputation

74 Good

About jovial_cynic

  • Rank
    Senior Member
  • Birthday 08/10/1978

Contact Methods

  • Website URL
    http://newprotest.org

Profile Information

  • Gender
    Male
  • Location
    Tumwater, Washington
  • Cars
    SR20DET 510
  • Interests
    welding, computers, building stuff.
  • Occupation
    Allstate Field Sales Leader
  1. jovial_cynic

    What did you do to your dime today?

    Pictures from the post above. It's not pretty yet. Most of what I've always done is just proof-of-concept, and then I end up sticking with it. The 4 lower gauges are all from the R33 skyline. For some dumb reason, I thought it would be a good idea to try to use it for my SR20DET install. The oil and fuel work just fine, but the SR temp sensor isn't compatible w/ the R33 gauge... so I have a temp sensor from an RB25 in my engine now. AF sensor is there on top. The dash is going to be built up to have the above instrument cluster tilted slightly towards the driver. I'm not sure what I'll install in the old instrument location or in the glove box location. Maybe a glove box? If so, I'll have to move my instrument cluster a little to the left. So... the only reason I wanted the R33 gauge cluster is because it had a built-in boost gauge that reads up to 7psi. I just needed the RB25 boost/map sensor. But it turned out that the boost gauge on the cluster I bought was BROKEN. So that sucked. Not to be undone, I had a 5v gauge sitting around, and a generic/GM 2-bar map sensor was cheap and easy enough to obtain. The GM Map sensor reads as follows. This is perfect. I did some image editing and made a background for the 5v sensor that correlates w/ the PSI and inHg. And voila. Boost/MAP gauge. I just need to build up a housing for the gauge so it fits w/ the rest of the random theme of my car.
  2. jovial_cynic

    What did you do to your dime today?

    Took it out for a spin for the first time in about a year. I decided to take a break from my massively complex telemetry system and just put it back together. However, I needed some gauges, so stripped the R33 gauge cluster I had been using, bondo'd up a new instrument cluster to put the tach, water temp, oil pressure, fuel level together, as well as leaving a spot for the AF gauge. On the trip out, I discovered that the cotter-pin holding the brake pedal up had come loose... and right as I was turning into Home Depot to grab some stuff for the house, the pin fell out and I lost my brakes. Fortunately, I was already in a parking lot and could down-shift myself to a stop. That was scary. Turns out Home Depot carries cotter pins.
  3. jovial_cynic

    Arduino + Raspberry Pi = 510 Telemetry System

    I gave that some thought. However, I've got another 20 wires that I have to solder onto the board for the JST clips, and I'm going to put those under the board. That way, I'm keeping my circuitry design and my sensor wires separate from each other.
  4. Between upgrading all of my brakes (discs all around), fabbed up new intercooler plumbing, and finally dumping the R33 skyline gauge cluster, I decided that the next major project to tackle would be the big electronic project: the Arduino/Raspberry Pi telemetry and custom gauge cluster system. It doesn't sound like much, since I haven't seen many overly complex Arduino or rPi builds, but this has been a pretty significant undertaking. The basic premise of this project is this: Use the Arduino to pull data from no less than 16 sources. On a standard Arduino UNO, this is not possible; there are only 6 analog inputs. However, using the magic of multiplexing, there's a way to quickly poll multiple sources over a single analog input; with two 8-channel multiplexers, we get that 16, plus the remaining 4 on the board itself. Once the data is pulled (either direct 0-5v values or values drawn from a voltage divider circuit), it is neatly organized into a (name:value) pair, and sent as plain-text serial data via USB into the Raspberry Pi. On the rPI, I'm using Perl/Tk to pull in that serial data and display it graphically on a custom instrument cluster as well as saving the data for future analysis. And then all of that is displayed on a 9x5" LCD screen. Easy, right? Well, then there's the whole project of packaging it all up. So, that's where the 3d printer comes in. I assembled an ANET-A8 3d printer, learned how to use Blender, and printed out some bits to bring it all together. The first thing I needed to print was an enclosure for the LCD driver. It came as a PCB board attached to the LCD screen with a power supply line and a ribbon cable; boxing it up meant printing a custom enclosure that I found and modified on thingiverse.com. The model online didn't include a lid, so I used Blender + my 3d printer to build one. Fits like a glove. Perfect! The next thing I needed to do was to figure out a way to mount the LCD. It's going to sit in the Datsun 510, so I can't just have it sitting on the dash. I printed these corner brackets, and they seem to work just fine. It's not beautiful, but it is very functional. Again, using blender to design it. I've used the bracket in several of my 3d prints, as it's perfect for small screws. Fits the LCD screen. It's a little loose, but maybe I'm compensating for heat distortion? Test fit. And screwed down to a piece of plywood. The next project was to build a custom Arduino. Because I have a need to take things apart and put them back together, I wanted to do one of those custom "minimalist Arduino" projects where I just use the atmega chip, a few resistors, a crystal, and some capacitors, and voila - homemade Arduino. Here's a standard arduino UNO. It's the one I used to test all of my code. And here's the Arduino connected to the two 5041 multiplexers. Perf board with the chip plugs. Always use these things; soldering down (and then trying to remove) a chip sucks if you have to re-orient later. Here is my minimalist Arduino board, together with the two multiplexers and a USB interface board. In case you're nerdy, here's the multiplexer pintout. And here's the Arduino atmega chip pinout. With 16 inputs, I didn't want a bunch of loose wires everywhere, so I decided to use some JST clips to organize my wiring. I couldn't find any 16-terminal JST plugs, so figured having 4 sets of 4 would work just fine. This is a 4-wire JST clip set. Not satisfied with having the clips dangling about, I did some test printing and built a holster for the clips. Perfect fit! However, this holster was just a proof-of-concept. I wanted to make sure they would fit properly so I can build an Arduino housing model that would incorporate the design. And here's a shot of the prototype of the Arduino housing. You can see the brackets that I've reused from the LCD clips, as well as the JST holster. And the arduino+multiplexer perf board fits perfectly. And here's a tall lid for the enclosure. At the time of this writing, I'm still trying to figure out how to pull it all together neatly. Right now, I've just thrown it all onto a piece of plywood. It works, but this is not likely the final product. On the left is the Raspberry Pi, connected by HDMI to the LCD driver housing in the center. On the right is another project - it's going to be the power supply for the whole thing. The LCD screen requires a 12v power supply - no problem. The Datsun 510, like most cars, has a 12v battery. I'll need to include a voltage regulator circuit to address voltage spikes and reversals, since that'll destroy my electronics. The Arduino and the rPi, on the other hand, only need 5 volts, so I have a switching converter to create a 5v power supply. This is particularly useful because several of my sensors need a 5v source as well: the MAP sensor (boost sensor), for example. The Arduino also accepts 0 to 5 volt inputs, so building voltage divider circuits from here is also important. Here, I have the Arduino hooked up and ready to go. The Raspberry Pi is loading... And we're in X-Windows on the Raspberry Pi. And here's the Arduino IDE serial monitor, and it's correctly reading the values from the rPi and multiplexers. There's no sensor data at the moment, so all the inputs read high. They would read low if they were grounded, and they would read correctly if they were connected to sensors properly. And lastly, here's the custom gauge cluster I coded. The tachometer is from an image of a Nissan Skyline tach; the Fuel Level, Oil Pressure, and Water Temp gauges were just cobbled together from some generic images online. I drew up the Boost gauge, and it may change to fit the scheme better. Once I have some data coming in from my accelerometer, I'll have the "g-force meter" working properly as well. This is all done within Perl/Tk. Stay tuned. More to come.
  5. jovial_cynic

    280zx struts

    I had a hunch that I should probably release a bunch of the pressure at the calipers, so I did that, re-bled the lines, and I think we're good to go. I'll drive it around and see if I notice any issues. Thx!
  6. jovial_cynic

    280zx struts

    I just checked; the brake pedal had zero play, so I was very excited that I was going to be able to fix the problem. I fiddled w/ the turn-buckle end of the rod, and I adjusted the lever-stop to give the necessary travel room. No change to the brake situation. I disconnected the push rod entirely to see if that would cause the brakes the loosen up a bit... and it didn't. The wheels move by hand, but there's enough resistance at the pads/caliper to prevent the wheel from free spinning at all. Other ideas?
  7. jovial_cynic

    What did you do to your dime today?

    ha! FB does tend to bring out the trolls. However, I ended up liking it better behind the grill anyway. I finally drove it around yesterday. The SR motor has been a little clackity clackity on deceleration, and after replacing lower bearings and verifying oil levels, and I think the noise might be a bad chain tensioner; I'm going to work on that at some point in the future. Immediate projects: 1. Get the brakes figured out. I've got discs all around now, and they're not releasing all the way. Also, need to bleed them again. 2. Install some gauges in the car again. I've been running all of my sensors through an R33 skyline dash... which I thought would be cooler than it actually is. So I pulled it, and now I have a bunch of sensor wires dangling in the car. No tach, no speedo, no temp, no pressure... just a bunch of hope that everything is running ok! I'm going to run everything through a digital gauge cluster I'm building (arduino, raspberry pi, etc., etc.)
  8. jovial_cynic

    280zx struts

    Ok - quick rundown: 1. Put in larger maxima brakes. 2. 280zx MC. 3. Put in rear discs Problems: 1. Brake drag. Can you adjust discs like you can drums? Is this likely a MC issue if ALL the brakes are grabbing? I lifted the car in the air, and every caliper is gripping. 2. The brakes are... squishy. I'm pretty sure I have more air in the lines, so I'll spend some time bleeding them again. But I really would have expected the brakes to be more responsive. Stopping on a dime and all that. Is there any chance that the air in the lines and the grabby calipers are related? Once I've bled the lines again, where should I start for diagnosing?
  9. jovial_cynic

    What did you do to your dime today?

    I installed my intercooler. I originally put on the outside of the car, but got some flack for it on the FB group, so I reinstalled it on the inside.
  10. jovial_cynic

    What did you do to your dime today?

    Rear discs installed. Front brake upgrade done. New radiator installed. Just need to bleed the brakes and fab up some intercooler piping and max out my turbo while the weather is cold.
  11. jovial_cynic

    280zx struts

    MC swapped out, pushrod addressed. Next step is to replace the struts left-to-right to avoid the sudden death-by-break-failure issue. I'm about to do the rear disc conversion on the 510, but I have to order the parts. I just need to wait until next pay day. Sigh. I blew the budget on boat expenses, which are quite a bit more than car expenses. I'm tempted to run without the rear discs, just so I can get the car back on the road sooner than later. But... on the other hand, I have a radiator swap to do (which means I have to re-located my turbo cooler source), as well as install the intercooler (which means I have to fabricate some new plumbing), so it's not like the car is going to get "done" any time in the near future.
  12. jovial_cynic

    What did you do to your dime today?

    280zx MC swap done. I've already swapped out the front strut/brake assembly, but it looks like I need to swap left/right assemblies to avoid having sudden brake failure. Next stop: rear disc brakes.
  13. jovial_cynic

    280zx struts

    And the push-rod question?
  14. jovial_cynic

    280zx struts

    I've FINALLY gotten around to doing this brake swap. I've been sitting on a set of Maxima struts since April 2006, according to my project blog, dragging them along with me to California when I moved down there, brought them back up to Washington when I moved back here. I've had a bit more time to wrench on things, so here we go. The 510 perches have been cut/welded onto the 280zx struts, new strut inserts installed, everything sealed up and installed. And THEN I read this thread about swapping left-to-right, so I'll get those flipped this week. I haven't addressed bleeding the lines yet, so this should be fairly easy. I picked up the brake master cylinder for the 280zx. Quick question: is the push-rod interchangeable? The 280zx didn't come with a push rod, so I hope it's an easy swap. If not, is there a known adjustment or tweak that needs to be made? Next - I see from some research that the 280zx MC has check valves that need to be removed if running disc brakes all around. That's my plan. Is there anything else I need to be aware of?
  15. jovial_cynic

    What did you do to your dime today?

    Just welded up the 510 spring perches onto my 280zx struts. I forgot that I hadn't ordered the strut inserts, so I hopped into rockauto.com and ordered a set. Now I get to hurry up and wait.
×

Important Information

By using this site, you agree to our Terms of Use.