RonsRed88

I thought it was about time I started putting something in a work log since I've made so much progress. Here's one of my first posts for a little background information:
http://www.mp3car.com/vbulletin/newb...time-work.html

So where do I start? I have most of the hardware and have configured the software. Here's what I have so far:

Motherboard Intel D945GCLF2D mini-ITX
power supply M4-ATX
radio Directed dmhd1000
radio cable
amplifier (AMP9 Basic from 41hz.com)
reusing existing speakers initially (Bose)
memory 2 GB DDR2 Crucial CT25664AA667
hard drive 320gb seagate
SD reader tribeca
wireless 802.11b/g PCI
blue tooth
gps DeLorme BT-20
panel mount USB port/cables
OBDI cable
(No head unit)

I have a lilliput 7" EBY701 touch screen on order and It should arrive tomorrow or Friday

For the front end I'm using RideRunner and I've got to tell you it's a very steep learning curve initially but the results are worth it. I've setup the music, HD radio, GPS, traffic cams, gas prices, OBDI software & cable, bluetooth phone, and internet.

I'm running XP, street atlas 2010, Blue soleil bluetooth software, Windows Media player, and TTS Datamaster (for OBDI)

This project is testing all my skills and making me learn new ones.

One of the first things I did was buy the motherboard, hard drive, cheap power supply for setup and testing, and yanked the DVD drive out of my home PC to load the software. I already had a spare monitor, keyboard and mouse so after that I was in business. I've intentionally kept the software as light as possible so the boot time is pretty fast. It takes about 30 seconds flat from power on to music playing from hibernation.

One of the hardest things for me to find was a small enough amplifier. I ended up buying the AMP9 Basic kit from 41hz.com. This was my first time soldering electonic components so I definitely learned a new skill (and bought some new tools!). It was a fun project in itself.

The HD radio initially gave me some real problems. When I played my mp3 player through the AMP it sounded great, but when I ran the HD radio through it, it was extremely loud. I had to turn the PC volume all the way down and it was still loud and sounded like there was a ton of interference or distortion. There was also a lot of noise from the PC going through the AMP. I eventually learned that there is a volume command for the HD radios in Ride Runner. As soon as I turned that down the HD radio sounded great but I still had a lot of PC noise. It kind of sounded like R2D2 humming to himself. I put a ground loop isolator in between the PC and the amp and voila life is good!

So then it was on to building the case. I decided to make my own case to ensure it would fit in the 1.5 din opening in the dash. The best thing I did was to mock it up in cardboard first. Getting all the parts to fit in such a small space takes some thought. The PC and power supply take up most of the space behind the radio so I'm going to have to figure out something for the amplifier and radio. I'll add some pictures in my next post.

So that's where I am currently. My next big challenge should be the fab work on the dash. I'm looking forward to playing with my plastic welder.

MP3Car.com

 

RonsRed88

So here are some pictures:

Motherboard


components strung out on the desk during testing, there is a desk under there!


Stock Dash: (some people call it the atari dash because it's all digital from the 80s)


This is how much room I have to work with behind the factory radio. That back brace will have to go.


Mocking things up in cardboard:


Mockup showing power supply and case fan placement:


I'm going to have to shorten up those power supply cables:


Test fit, it's going to be cozy!


Case Fab work started, so far so good!


I have to admit, I think it looks pretty cool:


I screwed down the hard drive and cut out a notch for the power supply to poke through


So I have to create the lid still and cut holes for ventilation. The power supply will be mounted to it as will the case fan.

Now you are up to speed. Time and money permitting I should post some more progress here soon.

RonsRed88

I received my Lilliut EBY701 monitor! I tested it to make sure it works, configured it for auto on using the directions in the FAQ, and then quickly disassembled it to see if it would actually fit in my dash and work when in pieces. So far so good! I took several pictures inside and out in case you are curious about the inner workings. Here goes:

From the factory running RideRunner:


Close up of RideRunner:


Cracked open the case:


Started taking it apart:





Measuring the heighth of the screen without the case:


Measuring width:


It still works!

RonsRed88

As you may have noticed in one of my previous posts I had to shorten the cables on the power supply (24 pin connector). Originally I thought I'd have to cut out a section of the wires and butt splice them together. Luckily I noticed Fry's has some pins and tools for working the molex connectors. I bought some pins and a crimper and did a test on one wire. It took me what felt like forever to remove the wire from the connector. I did some searching on the internet and found a tool for about $25 for removing these tiny connectors. It looks like a very simple tool. It's basically two tiny prongs sandwiched around a small spacer that allows the prongs to be inserted along side the pin depressing the tabs that keep it from backing out. You shove the tool in, push the wire in a bit and then pull the wire out. I figured I could make my own since I had some scrap sheet metal from my case. Here's how I shortened my cable, step by step:

Step 1, cut the cable. There's no turning back now!



Step 2, strip 3/16" from the cable, repeat 24 times!
Step 3, crimp on the connectors


Step 4, get the old wires out of the connector (or buy a new connector). This is where my home made tool came into play:



Step 5, start reassembling: Stick the pins in the connector. I hope you paid attention to how they came out!

Half way there:


Finally done!


The test - fits good and works good!

RonsRed88

I think my M4-ATX PSU is afraid of the dark! I haven't been able to come up with any rationale reasons so I'm entertaining all ideas now.

Anyway, here's what's going on. I finally got back to finishing up my case and built the lid and put ventilation holes in. I mounted the PSU on the lid upside down and turned it on with the lid still off and it worked great. Then I put the lid on and turned it on, sort of. It turned on the computer's fan and within seconds shut back off. Hmmm, maybe I forgot to hook something up. So I took the lid off and didn't see anything wrong so I tried again with the lid off. It worked perfectly. The computer booted and when I turned off the "ignition" it hibernated and then turned off the power. I'm testing on my bench using another power supply to simulate my battery. I've tried a lot of different things and if I leave the lid cracked open a bit it works, but then I close the lid completely, it doesn't. I thought maybe the mounting screws were causing a short and so I switched them out for nylong screws. Nope! still has the same problem. I tried putting electrical tape around the lid to prevent metal to metal contact between the lid and the case and tested it with my multimeter to make sure there wasn't any contact. Nope! still has the same problem. I rested the lid on top with the metal touching and it worked perfectly. It just won't work with the lid closed. I thought maybe the wires get bent a little too far with the lid closed, so I ran it with the lid open but reached in and pressed the wires down as if the lid were closed. It worked fine. I'm hoping to have an Ahah moment, but so far all I have is an oh damn moment. Any suggestions will be welcome. Here are some pictures:

Finished case from multiple angles:





Working fine with the lid open and metal touching:


Doesn't work with lid closed even with electrical tape all around:

rickhigginshtbr

damn you, now I may do this in my 85 once I'm done the one for my GA...

__________________

rdholtz

I don't see the P4 cable in there; that would mean no processor power and it would produce that quick shutdown you're experiencing (I did that myself; same board, same PSU); if I just missed it, is the P4 cable loose? Or is there a wire -- possibly in the shortened ATX cable -- that's failing contact when stressed by the bending when the lid is closed? (I don't care for crimped connections; they're flaky, and are the most likely culprit.)

That's very interesting sheet metal work, and nicely laid out. I especially like the hard drive tabs.

A couple of suggestions:

• The screws and nuts holding the fan on are dangerous: the fan is a source of vibration, and getting regular fan screws -- that cut their own threads into the plastic fan body -- eliminates the possibility of a nut loosening and falling onto the mainboard.
• Add more or bigger holes for ventilation air to move through, especially at the fan. There's no such thing as too much airflow.

Nice work!

__________________
.
If just enough is really good, then too much ought to be perfect.

2006 Scion xB with in-dash Atom & Lilliput 889GL -- Worklog at http://www.mp3car.com/vbulletin/work...res-links.html
.

RonsRed88

Thanks for the ideas. I'll triple check that 4-pin connector again. I wasn't aware there were special screws. I'll look for them. You are right about the vibration. The car vibrates enough as it is since the suspension is so stiff. I don't need any nuts falling off!

Ron

MP3Car.com

 

RipplingHurst

What he said ^^^^

I see the yellow and black p4 cable on the first picture, but I don't see it on the last one!



But it's gone here, after your cable rebuild:



Thing is, it still works with the lid open? It shouldn't! So okay, another thing I would think would be the two fans, one stalling the other, or more likely, heat right over the RAM banks, or maybe the wires are displacing the ram slightly. Try removing the M4 from the lid and place it where it should be with the lid open. Now turn it on, see what happens.

Last thing, it's cooling issues. Maybe it's getting too hot in there, not for the processor, but for the RAM? You definitely need more holes.

__________________
Worklog - 07 Infiniti Fx35 Carputer

Worklog – 06 Infiniti M35 Carputer (sold!)

rickhigginshtbr

You gotta keep in mind these C4's are pretty warm cars too, last year I didn't put it away til thanksgiving and was still driving with the roof off.

__________________

RonsRed88

I looked back at the picture I posted and realize I posted the wrong picture. I do actually have the 4-pin wire hooked up, but not when that picture was taken. You may have also noticed the hard drive power isn't hooked up in that picture either. It definitey wouldn't have worked in that condition! Good eye guys!


I plan on doing a couple things to test.
1. I'm going to turn that case into swiss cheese with holes
2. I'm going to build a clear plastic lid so I can see what the heck is going on inside.

After the previous suggestions I reterminated both ends of the 4-pin cable with the same results. I'm reluctant to do it with the 24-pin cable (too lazy), but may do so if all else fails.

Ron

rdholtz

You might hook a speaker or a set of headphones to the Audio Out port, and listen for beep codes.

Since it works with the lid off, try starting it with the lid off and the P4 cable disconnected. See if you get the same condition as you do with the lid on and the P4 cable connected. If it's exactly the same result, it could still be the P4 cable.

Or maybe not . . .

Try turning the P4 cable end-for-end to see if one of the connectors is flaky and reacting to the bending of the cable. Do the same with the ATX cable.

Could a cable be pushing the memory out of place?

Next, use some scrap sheet metal and build a half-lid. Mount the PSU just as it is on the full lid (I wouldn't bother with the fan -- unless it's dead shorting or moving something, it probably isn't the source of your problem). Then you'll be able to see in through the open back half and see if anything looks odd.

Just increase the lid holes where the fan itself is. Don't drill any other holes in the lid. That keeps the air blowing out of the fan from going right back into the case. It's best to keep the inlet and outlet as far apart as possible so you get good consistent flow. Plot a path from your inlets to your fan; the bulk of the airflow should be over the components that are the greatest heat sources. Make sure the air has a place to go when it's installed in the dash; you may even want to position a fan in the dash to blow air onto or away from the PC.

__________________
.
If just enough is really good, then too much ought to be perfect.

2006 Scion xB with in-dash Atom & Lilliput 889GL -- Worklog at http://www.mp3car.com/vbulletin/work...res-links.html
.

RonsRed88

That's one of those, "why didn't I think of that" suggestions. Great idea. I tried it and had no beeps.

So I decided I needed to see what was going on inside and built a plexiglass lid. It turned out great and if I weren't concerned with RF interference from the PC I'd keep it.
Here's a picture:


I clamped the plexi between two pieces of wood along the bend lines and then clamped whole thing in my vice. I then applied pressure while I moved my plastic welder back and forth along the bend line. As soon as it got to the right temperature it bent and then I let it cool in place.

Interestingly, the computer still acted exactly the same as with the metal lid!

This tells me a couple things. The metal lid wasn't causing any short, the wires aren't getting bent out of shape, and the memory chip isn't getting displaced.

I did another test and started the computer with the lid off and then slowly put the lid on. As soon as the lid was on the computer shut down. But it gave me a clue. The computer actually started hibenating before the power shut off completely. This made me think something was going on with the on/off switch from the PSU. So I removed the wire from the PSU to the mother board on/off pins and used a separate manual switch.

Voila, the computer runs! But I have to turn it on/off manually.

So that confirms the cables are good and airflow isn't causing the problem (although it still needs better airflow)

I ran out of time last night but I plan on taking a closer look to see if there is something shorting the on/off line to ground on the PSU. There is a rivet in the vicinity. I'll drill that out and see if it is the culprit. It might even be the metal case itself. One of my coworkers noticed from the picture that the case may be bowing when the lid is on. Perhaps it's touching somewhere.

RonsRed88

I appear to have the M4 working stably now, but I can't explain why exactly. Since my last post I tried a bunch of things. I removed the rivet, no help, I moved the psu away from the walls at least 1/2", no help. I cut out more of the case around the battery cable connections, no help. So finally I started thinking it was either a problem with the PSU or my shortened cable. The FAQ on the M4 has a lot of old posts where people upgraded their firmware and solved similar problems. I had no idea what version my firmware was so thought maybe I was shipped an old unit. I couldn't find any appropriate cables at Fry's to connect the USB so I could run the M4ATX utilities. I ended up having to make my own. It isn't pretty, but I just wanted it for testing. Here's a pic:


Now here's the surprising part. I thought I'd have to jump through some hoops to get the computer running long enough for me to run the utilities, instead it booted right up, even with the metal lid on! The utility says I have version 2.0 of the firmware which surprised me because I thought they were only up to version 1.3


I went back and forth about half a dozen times removing and connecting the USB connector from the motherboard and it never failed to start with the USB connected. It always failed with the USB disconnected.

I'm not thrilled with the prospect of losing one of my 8 USB ports but I might be able to get away with it. I'll have to take a look at what I plan on connecting and make sure 7 ports is enough.

Now I just have to make a cable that isn't so hideous looking!

This power supply problem derailed me for so long I have to regroup and remember what I was going to do next. I'll keep you posted.

Ron

RonsRed88

I guess I'm becoming a little obsessive about cable management since there are so many wires in such a small space. The 18" SATA cable was definitely in the way. I was seaching on the internet looking for a shorter cable with a right angle connection facing the right way and I stumbled upon this youtube video
http://www.youtube.com/watch?v=H_RT-5Mgluo

It's one of those simple cheap mods that's just plain elegant.

Here's a picture of it in my case: ( I may have to edit this later, I appear to be having problems uploading pictures right now )


The next thing I did was take your advice and increase the airflow in the case. I enlarged the exhaust holes in the lid and added a lot more holes in the sides of the case. I was using the M4 ATX monitoring program to see how hot things were getting and before I added the holes it was getting up to 50c and now doesn't go above 32c. Of course it's still sitting on my desk and isn't nearly as restricted as it will be in my dash.

Then, I hooked up cables to power my llliput from the M4. I originally was going to use an M sized power plug as suggested in the FAQ and the cables I was going to use were just to big to solder to the plug. So I was looking around for some smaller wire and realized, I'll never use the cigarette lighter adapter that came with the lilliput. So I cut off the wire and connected the bare ends to a molex connector and plugged it into the power supply via one of the open hard drive plugs. That worked great and now the monitor starts up when the computer starts up.

MP3Car.com