Sunday 29 December 2013

KAW RF - Adding 2x 100nF caps to HV circuit

Based on Mike Tranchemontagne's advice, I had previously added a 100nF to the High Voltage circuit of the KAW in order to boost the current to supply the RF circuit hack. This gave the results seen in the previous post.

 

The v_in was stuck at about 2.4V drawing about 56mA never reaching a higher voltage in order to activate the 3.3v regulator.  In order to see more energy and perhaps climb over this 2.4V I added another 100nF capacitor in parallel. In total this is 670nF (470nF + 100 nF + 100nF) which is a 43% increase in energy storage capacitance!

 

Start up the oscilloscope, and plug in the KAW! It's not the smoothest but after a few seconds, the circuit does achieve its goal to power the moteino microprocessor!


1) 0.0 to 12.5 (12.5 seconds) output is about 2.3 V. The KAW screen is completely dark during this time not displaying anything.
2) 12.5 to 21.3 (8.8 seconds) output is about 3.3 V. You can see the 13V9 line start to increase, and this is not causing a proportional climb in the v_in. The KAW screen is now showing the correct voltage on screen.
3) After 22 seconds the output finally reaches about 5V for steady operation on both the KAW and the moteino.

3 comments:

  1. I just swapped my 470uF with 1000uF on the low voltage circuit to see how it would affect the voltage graph and timing. It still took 20 seconds to fully initialize, but only took the 13V9 line about 12 seconds to fully initialize, and thus the KAW screen was displaying correctly 8 seconds before the moteino booted.

    ReplyDelete
  2. Added yet another 100nF to the HV (Practical mod will be to use one 0.33uF in the future). In only 2 seconds the 13V9 line becomes stable with KAW. The v_in output will charge and idle at about 3.5V for about 11 seconds until then climbing to 6V. This is a great improvement, and in practice the KAW user will only see a dark screen for 1-2 seconds, and I can modify the code to wait 15 seconds or so before getting out of an idle low power state.

    ReplyDelete
  3. While modifying the code to add more of its intended functionality, I also added two SLEEP_8S delays to boot up as previously mentioned. Checking out the voltages, it only made the climb to 6V that much longer. Modifying the code to only boot up with a SLEEP_4S delay instead had the opposite effect of speeding up the whole process. Now I understand! .. It only takes the initial 2 seconds to power up the KAW and moteino to its low power stage in full functionality which needs only about 3.5V peak. Once the code takes place, it will wait the 4S until processing the blink/RF code which demand the extra energy from the KAW and fill up the capacitor to 5.5V! Success!

    ReplyDelete