Friday, August 19, 2016

Laptop mainboard - ALMOST right

My last post was about a laptop mainboard that I dropped in, and that it only had one thing different - a different connector for the feature buttons.

Well, then I noticed that the touchpad buttons didn't work.  And it didn't think there was a battery plugged in.  And the LEDs up front (power, battery, wifi) weren't lighting.

I opened it up again, and found that there were no switches installed on the mainboard for the mouse. I surmise that it was for a slightly different model that had buttons in the touchpad (there is also a populated connector on this mainboard in the right place, which is probably for that purpose).  So I got the hot air out and removed the switches, and moved them. I also desoldered the LEDs from the old board and moved them (no mean feat - they are 0603 LEDs).  I plugged in enough to power up, and no luck. The LEDs didn't light, and the mouse buttons still didn't work.

I compared the boards and determined that the mouse button switch leads go through a couple of 0603 100 ohm resistors. One was still on the board, so I moved that one. The other one was probably lost when I swept the switch off with the hot air, so I put in a 330 ohm resistor which I have on hand. This worked.

The LEDs, probably a similar story because I noticed that it had LEDs already in place elsewhere on the board and those were working. I just drilled three holes in the top of the laptop to let the LEDs shine through and covered it with tape.

The battery - that was another story. Turns out the battery connector was about 3mm shorter on this board than the other one. I did manage to remove the connector from the old one and get it on the new one. This took over an hour. Lead-free solder is a pain in the ass when desoldering large connectors with through hole physical stakes.  On the old board, I ripped the via rivets out while desoldering, so I was super careful on the new board. I tried hot air on the old board first but it was just not working and I decided if I was going to have to use other means on the new board, best to practice on the old one first.

Anyway, that's all working now, the LEDs are shining through the lovely bodged holes drilled in the case, the battery is charging, and the mouse buttons work.

I would NOT have been successful at this stuff without the microscope that I bought last week. The parts are just too small for my eyes.  I think they're probably too small for anyone's eyes when you consider you need to examine the solder after your work is done. Maybe you could do it with a 10X loupe but I'm happy to have the microscope.

Monday, August 15, 2016

Laptop repair

Just a quick heads-up - we had a laptop die in our household a few months ago, and didn't have time to mess with the repair so we just bought a new one. Last week I started messing about with the repair and spent several hours looking for a crack in the circuit board or a bad connection to a component (would only power up if pressure was applied to the board).

After a couple of hours of this, just out of curiosity I checked ebay to see what a new mainboard would cost. Turns out, $30.  Well to heck with fixing the old mainboard then. I'd already wasted more time than that.

2 days later, the mainboard arrived. It was new in box, slipped right in with one exception - it had a 0.5mm pitch connector instead of the 1.0mm pitch connector in one place but it was just for the options keys.  I ordered new connectors off eBay, it'll be a few weeks (China) for $3 for 5 of them.

Moral of the story, check the prices for all the options before spending much time on a repair. I just assumed mainboard replacement would be uneconomical since the last time I tried to buy a laptop mainboard they wanted more than the cost of a replacement laptop. These days if the laptop is more than 2 or 3 years old, likely there are repair depots liquidating excess inventory for a song.

Sunday, June 12, 2016

First Shapeways order

I did a 3D model for a friend and it needed to be printed in metal, so I sent it off to Shapeways.

The results are, it's not bad but definitely not up to being used as a replacement for machining, but I think we will be able to use the piece, after some rework.

There's a small vent hole in the top which was closed when I received it, but it appears it just wasn't broached well, a small drill cleared the printing powder easily.

The OD of the original was 14.18mm.  The printed object is around 13.9mm.  The ID was set at 6.09mm, the printed object varies from 5.95 to 6.09mm.  So in general not bad at all.

The surface is fairly rough as you can see in the photo. It'll probably do for this application, though.

The material is quite hard. Drilling it out round is pretty difficult.

Monday, April 4, 2016

Sneak Peek

Couple of photos from a fairly interesting project I'm working on. It's taking more time than I bargained on but that's mainly because I don't like doing half-assed jobs at things.

I'll be doing a full post on this in a week or so after it's delivered.

Friday, March 18, 2016

FRC Stronghold videos

This is just a list of all the videos I find of FRC bots - mainly looking at designs, but also good field video as well. I'll be returning to it and adding links frequently. I intend to use this as an off season learning tool.

Reveal videos and detail photos/video

Related videos

Chairman's videos:

Design photos
971 photos and here

Sunday, February 21, 2016

Phone fiddling

Over the last 3 weeks or so I have been battling what I thought was an Android issue on my LG Nexus 5, where the speaker on the phone would not always work.  Long story short, after about 10 reboots and 2 factory resets, I found that it was actually a hardware issue.  Pressing firmly on the back of the phone would make the speakers start working.

I opened the back of the phone (which is relatively easy) and cleaned the contacts.  This seemed to make the speaker happier but left the vibration motor unhappy.  I opened it again, stretched the spring contacts for the motor a bit, that fixed that but later I found that wireless charging wasn't working.  This evening I opened it up again, carefully cleaned ALL the connectors, used an exacto knife under magnification to ensure that ALL of the spring contacts were fully as strong as they could be, then ressembled.  At this point it seems like everything is working correctly.

I was thinking about getting a new back for it since the retention clips get worn when you open it, but the available backs do not have any of the electronics on them, they must be transferred, and there are I think 4 stick-on antennas and a large wireless charger pad, so this is a bit intimidating to get them all cleanly unstuck and re-stuck.  I'll leave it for now and hope for the best.

It was interesting to open the phone. I'm not happy with any phone that comes with a "non-replaceable" battery, but at least I found that this battery is actually very easy to replace and is available for $10, so if the phone lasts another few years I can drop a new battery in pretty easily.

Thursday, February 18, 2016

Ultrasonic sensor for FIRST robotics

I made an ultrasonic sensor for the FIRST team in town. We're actually modifying it a bit so I'll be posting another one later, but I'm posting this one now in case anyone else wants to use it.

It just continuously sends distance measurements (3 to 5 times a second) back to the RIO using serial.

Here's the Youmagine page: