Feed on
Posts
Comments

Today we’ll be taking a look at the Linksys PAP2 VoIP Phone Adapter which has an Ethernet port and 2 phone ports for analog phones, powered using a 5V adapter.

2 screw at the back, 1 screw on the side and we’re in.

We have a 2 chip solution – the main chip and LAN controller. The bottom of the board looks like it’s seen better days and there are some spots around the board that looks like burn marks, not sure if someone else has opened this up before. For the power side, there are 2 chokes before going through a PNP transitor to a 470uF capacitor and we have 2 AP1117 LDOs giving out 2.5V and 3.3V. There are 2 logic chips around as well, an LS32 a quad OR gate and a HC74 Dual D Flip-Flop with S/R. PCB date code is 40th week of 2010.

(more…)

Read Full Post »

From the previous part, we designed a small power supply based on a Richtek DC-DC, added an 128X64 OLED display and gave it a quick test which it seemed to cover most voltage ranges with better increments than the old SPPS.

(sneak peak)

The PCB’s arrived and we’re ready to start populating each segment individually to make sure each part works. I put on the ATmega328 with the LCD, tested ok. Next was the main DC-DC converter with digital potentiometer and to test controlling it from the ATmega which seemed to work fine. One thing I hadn’t really paid attention to was that the DC-DC converter only goes up to supply voltage – 2V or so from my testing, the reference voltage starts to drop from the nominal 0.8V and it started to oscillate, so I might need a 18V power supply if I want up to 15V output.

Output Enable troubles

Next I soldered the P mosfet for the enable so I could enable or disable the output, that’s where things didn’t go as planned. I hadn’t considered the case where the power supply was set 2V and the gate was low, Vgs would only be -2V which is not enough to turn on the mosfet, you need something like -5V or higher to full switch it on, basically under 5V it was unusable, the mosfet would be dropping most of the voltage.

(more…)

Read Full Post »

My small bench is located in one corner of the living room next to a sliding door so soldering fumes didn’t appear to be much of an issue when I had a fan blowing the solder away from me.

I thought it was probably time for a small upgrade so I bought one of the Solder Fume Extractor’s from Ebay, $24 locally, a bit pricey for what is it. Once I received it, it seemed to work fairly well, the distance you could be away from it and it still suck the fumes was around 15 to 20cm. If you didn’t want it that close, you could always use another fan to gently blow towards the fume extractor or use some materials to focus the suction of air.

After using it for 10 minutes, the fan started to smell and was getting a bit hot too. I left it running for an hour but still had the same issues, ugh, so it was time to take it apart and see what I can swap out.

 

The device has the AC power cord going directly into it, so I thought they might have a AC to DC converter with a DC fan. That wasn’t the case, upon unscrewing the 4 screws at the front, it’s an AC fan. Power is just running though the switch to the fan.

(more…)

Read Full Post »

Today we’ll be taking a look at the Netcomm NB6Plus4Wn ADSL2+ Wifi Router which is similar to the Netcomm NB6 Rev2 ADSL2+ we looked at previously except that this one has 802.11n Wifi and 4 LAN ports.

4 screw later and we’re in.

We’ve got a 3 chip solution, the main chip, LAN controller and Wifi controller which is using a discrete front ends for both antennas that look to have a balun in them. There’s a 5 pin header with only 4 pins soldered going to the main chip. Looks like there is also possibility for populating a USB header near the phone line port. PCB date code is 42nd week of 2011.

Something that stands out is the 2x Lelon 25V 2200uF caps which are swelling. We have a little bit of glue being used on the 2 inductors for the DC-DC’s as well as the one of the red caps to the line transformer.

(more…)

Read Full Post »


Wouldn’t it be nice to know it you have mail? I think it’s a good idea and had it on my to do list for a while. Our letterbox looks similar to the one above, you can place small items on the top, in the letter slot or open it up from the back.

 

At first I thought we could do this with a vibration switch however upon testing the switch it looks like it needs a fast motion in order to activate. Another option is a reed switch on the top part so if the mailbox is opened, it will detect that, but would leave letters undetected. And yet another option, is a light sensor placed inside the mailbox but also results in the same problem.


So one of our last options is to use an accelerometer to detect the slightest touch of the mailbox, most mailboxes are pretty rigid but ours has a tiny bit of moment to it, we can’t really put the device we would make in the letterbox itself as the RF would mostly be blocked so it would have to be placed outside. As usual, I’ll be using an ATtiny84 with the nRF24 to simply send a packet when the interrupt occurs.



The accelerometer I’m looking at as you can tell by the title is the ADXL345 which has 2 interrupts available for single tap, double tap, activity, etc, and current consumption looks to be pretty low. It supports I2C or SPI, we’ll be using 4-wire SPI.

So I tried the SparkFun ADXL345 example which worked well, they have all the interrupts available so all I need to do is take everything I don’t need out, convert it to C and try out the low power modes. Note that the ADXL345 is a 3.3V device so you will need a logic level converter if using it with a 5V Arduino.

(more…)

Read Full Post »

I have a couple of nRF24 devices around the house like the doorbell, front door camera, the single receiver for both and I might look to expand that list in the future but re-programming the receiver when you would like it to handle a new device or perform an additional action can take a while plus the time testing that everything works.

So I thought, why not make an ESP8266 for a web interface where we can store rules and have it connected to the nRF24 to receive incoming packets, it’s similar to IFTTT but all local. The rules would all be based on a single byte packet it would receive, we could have the rules active or not, with rule actions such as sending an email, pulsing a pin for a certain amount of time, sending an nRF packet, writing data to a common variable (with a timeout when it’s reset) and reply back with that variable. We can have a button to save everything to the EEPROM (emulated on flash) and read it all back if the ESP8266 is power cycled.

Some quick use cases:
– Doorbell is pressed, you could either have it pulse the buzzer locally for a second or two and you could also have a remote buzzer anywhere else so it sends a packet to have that buzzer go off too.
– I’m thinking about adding a small sensor to the mailbox, so if it’s moved, it would send a packet, we could have it write data to a variable. Add a battery powered sensor anywhere around the house with either a buzzer or LED, to check in every few seconds and read that variable back so if it changes to a 1, the buzzer or LED goes off for a little while.

(more…)

Read Full Post »

Today we’ll be taking a look at something a little bit modern, the Netgear Telstra Cable Adapter (CM450-1TLAUS) which is a cable modem with just the one Gigabit Ethernet port and RF connector.

4 screws later and we’re in.

We have a single chip solution with a large heatsink and the RF side has the usual RF shield. They have exposed ground strips on 3 out of the 4 sides of the board and there are no electrolytic caps to be found, they even had a footprint for one to the left but had the option of SMD caps too. Interestingly there are 2x 4 pin headers, labelly nicely, one reads UART0 (BBS) and the other UART1 (LINUX).

We have some logic chips around, an LV132 Quad NAND gate and on the bottom an 74HC74D Dual D-Type Flip-Flop and an LM358 too. PCB Datecode is 45th week of 2016.

For the power side, we have 4x DC-DC converters with inductors ranging from small to large, the largest one is the Richtek RT7233 18V 4A while the others are 6 pin chips (2x 04-P0G, 1x C8-L2P) which I can only assume might also be Richtek chips as the DC-DC’s don’t use an external diode.

(more…)

Read Full Post »

When finding out that one cell of my 3x 4S 1300mAh batteries were discharged to 3.3-3.4V (other cells were 3.7-3.8V) after not flying for 2-3 months, I was a bit surprised, these batteries were one of my first batteries purchased about 1.2 years ago. I’ve tried to be better now, checking packs every month, I’ve got some old packs for the RC car that hardly get used so I wrote down the idea to do battery storage voltage monitoring a few months ago so let’s make it happen.

I’m thinking of using the nRF24L01 module (as it’s pretty low cost), using an ATtiny24/44/84 which should give us 6S support and have it powered from the first cell of the LiPo, it shouldn’t take too much current from the first cell as we stay in sleep/power down most of the time. Add in a few resistors, led, mosfets to turn on when we want to measure the voltage. I will coat the PCB in lacquer and cover it with heat shrink to protect against the environment and since the board needs to be as small as possible, I’m going to try the SMD version of the nRF24.

Check-in times for these sensors might be in the range of hours as battery voltage won’t change too much, it would be nice to have a graph over say 30 days but I might look into that later if I feel it’s worth doing, for the moment it will just show the sensor number, cells, status if it’s ok or not, battery cell voltages and last check in time. Server side would just be an ESP8266 with the nRF24L01 directly attached so no need for any other MCUs.

(more…)

Read Full Post »

Today we’ll be taking a look at the Cisco DPQ3925-X Cable Modem / Gateway which is a cable modem with 4x Gigabit Ethernet ports, 802.11n Wifi, USB port and 2 analog phone ports.

4 screws later and we’re in.

Looks like we have a 2 chip solution, one of them has a pretty large heatsink. There is a Mini-PCI Express Wifi adapter with dual antennas. There is an RF can for the cable input and some conductive tape connecting it to the ethernet ports which is a little odd as the ethernet shielding is grounded. We also have tape near the main processor going to the memory and another small piece right next to the Mini PCI Express connector, strange. PCB date code is 37th week of 2014.

The main heatsink isn’t soldered down, instead the legs are bent so it can’t go anywhere.

Something else that stands out is that we have a 25MHz crystal which the case was soldered to ground.

The 2 Wifi antennas are located in the front of case.

(more…)

Read Full Post »

Previously I was looking at making a Dual Output Linear Power Supply, the prototype was working but while designing the PCB it would end up being large, a bit heavy with the heat sinks and I don’t think I would really need a linear power supply, it was a nice idea.

So it’s back to the drawing board as to what I really need, I do still have the SPPS which works but maybe it’s time for an upgrade and now that I know a bit more about DC-DC converters we can stick in the Richtek DC-DC chip instead of using an Ebay DC-DC module. We might as well switch out the 8 digit LED display for an 0.96″ 128X64 OLED SSD1306 display which I’ve seen a lot of projects move to.

The DC-DC converter I’m looking at using is the RT8292B which runs at 1.2MHz versus the 400KHz of the RT8293A which I was using previously, it looks to be almost identical except that it will allow us to use a smaller inductor.

Like we’ve done before, we will use a 10K resistor for R1 and 10K digital potentiometer for R2 to control the output of the DC-DC.

(more…)

Read Full Post »

« Newer Posts - Older Posts »