OPS - Energy Harvesting

1
OPS - Energy Harvesting

• Name Alex Phipps 12/13/2005
• Goals
• Mount solar cells to board.
• Test PZT with resistive and PRC loads to confirm
  bimorph models and experimental data.
• Key accomplishments.
• Most parts have been soldered onto solar board.
• Shengwens files have all been transferred and
  are functional.
• Key issues
• How to attach the solar cells to the board.
• PZT bimorph beams no longer have the same
  impedance when measured.
• Simulations are also difficult because the model
  doesnt allow for Ropt to be used.
• Next steps
• Populate both boards.
• Create test plan for new solar ICs.
• New design has test pins for charge pump.

2
Action Items

1. Build Charge Pump Board.
2. Conduct a literature survey to determine if the
   charge pump is publishable.
3. Construct board for solar converter/controller.
4. Plan a paper around it.
5. Simulate PZT converter with Shengwen and Anurag
   to verify 6uW with Bimorph.
6. Simulate resistive loading and compare with
   experiments.
7. Talk with Erin about gold plating/ Wirebonding
   from IMG.
8. More solar cells available?
9. Software from Shenwen and familiarize with his
   work.

3
Action Item Progress

• Build Charge Pump Board.
• On hold while solar board is being constructed.
• Construct board for solar converter/controller.
• Board milled and most components have been added.
• Simulate PZT converter with Shengwen and Anurag
  to verify 6uW with Bimorph.
• Simulations predict more than 6uW of power using
  the Bimorph model.
• Simulate resistive loading and compare with
  experiments.
• Resistive loading seems to have changed for the
  Bimorph beams since the first data was taken.
• Simulations cannot be easily altered to fit
  experimental results.

4
Action Item Progress

• Talk with Erin about gold plating/ Wirebonding
  from IMG.
• Probably will not be able to do it. Most likely
  will hire Al Ogden to complete the wirebonding.
• More solar cells available?
• IXYS has not responded to either of my emails. I
  also phoned in with little luck.
• Software from Shenwen and familiarize with his
  work.
• PZT and solar simulations finally both work on my
  account and backups are made.

5
Wirebonding

• Al Ogden will probably have to do this.
• Other options
• Gold plate whole board.
• Cost and still need to wirebond.
• Get gold plating pen system
• Requires time to get system and cost of pens.
• It will cost 52/hr and Al estimated 2-3hrs to
  do the whole job.
• There is also the cost of silver epoxy, which
  should be relatively cheap.

6
PZT Testing

• PZT tests conducted by Shengwen show the
  following

Pmax Propt R 15.84k Pp.r.
Calculated 34.3u 32.9u 9.65u
Simulated (in cadence) 34.1u 32.7u 9.86u
Measured N/A 13u 6u
7
PZT Testing

• The problem for the PZT is now this
• The model that Anurag has for the bimorph gives a
  Ropt 15.84k. Using this model in the cadence
  simulation with Ropt 15.84k, resulted in an
  output of 32.7uW. When the bimorph beam was
  physically tested, the Ropt 50k, and the
  maximum power output was only 13uW.
• It is possible that the R1 measured to make the
  LEM model of the bimorph had some error. Since
  R1 is a major component in Re, this could have a
  big effect on the Ropt. We want to edit Re so
  that Ropt is now equal to 50k. This means
• And since

8
PZT Testing

• We know that we must just change Re so that the
  impedance ZTH is 50k. However, due to the
  capacitance value that we found this is
  impossible.
• The measured Ropt was done rather roughly, so
  Anurag and I decided to re-measure.
• We found that Ropt had changed for the bimorphs
  and that the maximum power was no longer as high
  as it had been.
• This will require more investigation.