Electronics, Robots, Making and some random.
This is my first try at reflow soldering. It took four attempts to get it right… so you learn 3 ways not to do it 😉 I just got this 858D station in the mail the other day, it cost me about $70 on eBay.
)
In this video I referenced Jayson Tautic’s reflow example.
I pulled out the weather radio project today to see what I could get done in a few hours. I was pretty close on finishing off the hardware but I fell a little short right at the end. I found I had forgotten to buy something to convert the regulated 5V to 3.3V for the radio (and PIC since they’re tied together on I2C). I ran into a few issues I totally spaced:1. The Si4707 requires a reset after power up .. it ignores I2C if you don’t.
2. Pull-ups.. duh, not only on I2C which I had, but don’t forget the Si4707 reset (oops).
I ended up buying a couple random Digikey parts at the ham radio convention and guess what? 10@ MCP1802T-3002I/OT (300mA 3.0V LDO) ..they are SOT23-5 and I had JUST gotten 10 break-out boards in the mail so I ended up having a 5V3.3V converter (close enough anyways, as the PIC and Si4707 work down to 2.7V).. soldered it up but I didn’t have time to pop it in. I used the Weller but I think I’m going to solder another one with the new hot air gun I got this week. I ordered one of those cheap 858D rework stations; I don’t plan on using it too much so hopefully it’ll do the trick. I also got a bunch of SMD protoboards.. so I get to practice reflow this week.
Tomorrow I should be able to finish up the radio and then it’s all software.
The photo isn’t the most exciting workbench shot but you can see that 858D in the back corner. The WX radio is the black box right up close to the left.
I didn’t have a very productive weekend in the workshop. I received a ton of parts Friday and Saturday including the case to the WX radio. I started installing parts and pieces in the case. I’m always very nervous about the first drill hole.. A only made a minor error (so far) and it is nothing noticeable.
I spent much of the weekend with my wife doing little tid bits outside the house. She spent a lot of time home last week so she was sick of being home; that directly translate into a reduction of bench time 😉 She did go out for a work thing Saturday and I went down to the Mike and Key game radio swap meet. I purchased a bunch of odds and ends and gave up when my pockets were full and I couldn’t carry anymore.
This morning we went out for a quick bit after sleeping in a little and went for a hike out east. It was relaxing but ate up most of the day. I suspect I won’t end up with a lot of bench time until this next weekend: I will likely have a bunch of overtime at work as I just inherited another project when I’m already way overbooked. I know, better than not having work.
Some of my purchases include a cheapo hot air gun.. For reflow soldering. I don’t know if I made a mistake going cheap? I also bought a new power supply and just a ton of parts. I pulled my last 16pin DIP socket out and can’t believe I didn’t order some last time?! How embarrassing! I also ordered six of the new TAUTIC 18F26K22 dev boards on Tindie and some blog stickers.. Check out those dev boards if you’re into PICs, their pretty slick.
I’ll leave you with a photo of the falls we hit up on our hike today…
I was inspired by Alan Wolke’s ( @W2AEW ) video on measuring capacitors and inductors with an oscilloscope. I tried it out and it works pretty reasonably; yeah why wouldn’t it? Anyhow, if you can do it on an oscilloscope it can be done by a microcontroller right?  I have been trying to keep myself from buying this $220 eBay LCR meter on eBay.. it looks nice enough (model: MCH2811C). I needed to make this a project or I was going to pull the trigger on some Chinese garbage!
I went with simple and cheap, I don’t know how well it’ll work out yet but I bread-boarded a proof of concept design. It tested okay after some modification. The first road bump was I had found the PIC output pins had unacceptable rise time compared to 74HC14. The first change was using the PIC output to drive the 74HC14 HEX inverter to get the quick rise time needed. Â I’m throwing a fast edge at the tank circuit that includes an “unknown” inductor and then I do my measurement, just as Alan used his homebrew TDR circuit. I went really low tech on my measurement circuit, I may change this. I used an LM339 comparator and a trimmer as a voltage divider. The first couple waves in the tank “ring” trigger the comparator and I measure the frequency by figuring out the time between the positive pulses by timer. Pretty simple no? It works fine as it turns out. I will have to get a better capacitor and ensure I measure it very accurately to do my math in the PIC and get a reasonable result.
Instead of the usual photos I made another YouTube video. It wasn’t a great one, one take, no editing.. it’s gets my point across (kind of).
My project proof of concept for the PIC L-meter. With the addition of a known inductor and a rotary switch and a little more code you can turn this into a PIC LC-meter in no time.
I’m ordering stickers.. what kind of interest are there in these? Any? Let me know.. e-mail, tweet, comment..
EDIT: Oddly total silence despite the views.. let me restate this… any interest in the sticker if it was FREE 😉 Is that to obvious?
The sticker is 1.5×1.5 inch:
So as suspected the uLCD-43PCT display worked just fine once I got their USB-RS232 dongle.. it has a RST pin.. maybe it uses it in some way when programming? I didn’t do any research because I just wanted to try the display out but maybe I’ll check that out later.
I used Visi Genie which is one of four options you have for configuring their displays. It’s not bad.. but it’s not great. It gives you a very limited number of switches, buttons, numeric displays, etc.. but it worked. I also haven’t found a way to send text to the screen which I can’t believe I couldn’t do.. I just haven’t found it I hope? You can dig in and actually use some of the other configuration modes in which I imagine you can design your own display in detail…again maybe some time I’ll check that out. It’s all code and looks like it would make for a very long day or two. The Visi Genie took me about 3 hours to figure out and configure the display. I ran into a few issues.. of course some of my time was just picking what colors I liked best. The colors show up on the monitor a little different from the LCD so that took some tweaking.. also I noticed the LCD is a little less readable than their simulated display: no big deal though.
The really REALLY annoying bug that took me about an hour to figure out (while was figuring out the serial protocol) was the “LED Digits” module… I dropped it in, selected 6 digits with the decimal place in the third; check, no problem. When testing I could write 1, …. 2, ….3,… WTF? See photo below.
How did that happen? The serial commands were right… after a while I dropped in the Custom LED display.. this next photo shows both displays with “200” written to them.. note: the larger LED segments has the leading zero’s turned off (hidden), it’s setup at six digits.
…… sigh.
I finally switched my display from 6 to 4 digits as a test (well really 2+4 digits). Success! I guess the program doesn’t like over 4 digits which makes sense because after I figured the serial protocol out there is only room for 4 digits  in the packet. Why does the program allow for more than four? The program does a pretty good job of limiting everything else … oh well. It’s finished
Next steps … I’m deciding if I want to wait for the Tautic 18F26K22 dev board or order something and use one of my protoboards or even just stick with the 16F1509? I’ll figure this out shortly once I decide how I’m going to mash this all up together. I ordered another Si4707 from Richard @ AIW Industries. Hopefully his board is decent: It has promise with the BNC connector and you can buy a fairly cheap Larson whip from him for 12$ and change for the right band which is pretty nice (no more random-wire antenna!). I’ll probably have to shelf this project for a week until it arrives anyhow.
I started up MPLAB 2.05 after the update… I want going to start programming but didn’t get far before I ran into this..
It looks like it marks “TODO” in tasks!!?…. I don’t know if this was an option before but I usually just //TODO if I have something I need to get to… perhaps the update reset some toolbars? I don’t know but I really like this feature. I love me some MPLAB.