Microcontroller PI project questions

[Charles (Upstate NY)]

Okay keep in mind I'm a computer programmer not an EE and a PI newbie.

I have been eyeing the microcontroller PI projects for a few weeks now and I think I'm going to try building one this winter for beach hunting. I had purchased a Hammerhead board but being a computer programmer the microcontroller versions have captured my interest.

The thing is, I'm not satisfied with the reported depths for these published designs. I'm wondering if the power supply is part of the problem. The Gold Pic uses a 9.6 volt supply. My question is, would a 19.2 volt power supply offer me some options for increasing the depth?

I selected 19.2 volts because I want to keep the power supply reasonably light and utilize readily available 2900mAh NiMH batteries. At 19.2 volts I can use either 16 cell AA or AAA packs which weigh 16 ounces and 8 ounces.

The second question I have is related to noise and the EMI/RFI floating around the shop. I have heard many times that you have to keep the noise on a PI as low as possible so how do you measure this when household AC, the TV, wireless networks, etc. are spewing out all this noise?

Don't laugh but if you saw the movie Enemy of the State with Gene Hackman and Will Smith, Gene's character Brill had built basically a work cage surounded in copper wire mesh to create basically a space clean of EMI/RFI inside. What do you all do when working on this stuff to eliminate the external EMI/RFI?

Oddly enough there is a local supplier here in Albany who sells this sort of EMI/RFI protection, no doubt to poor people who think aliens are transmitting mind control waves at them or something. So I could create such a cube for testing or at least some screens to setup.

Well for now, that is all.

Charles

[bbsailor]

Charles,

Your Question:
"The second question I have is related to noise and the EMI/RFI floating around the shop. I have heard many times that you have to keep the noise on a PI as low as possible so how do you measure this when household AC, the TV, wireless networks, etc. are spewing out all this noise"?

If anyone has an oscilloscope that works up to the 200 to 250MHZ bandwidth, it is quite easy to see the effect of wideband EMI/RFI floating all around you. Typically, the oscilloscope display, when on the wide bandwidth setting will look blurry and then clean up when the bandwidth setting is reduced. This is the result of the oscilloscope probes and test leads picking up all these EMI/RFI signals.

Here are two practical tips when working on a PI design on your workbench.

1. Use a large metal enclosure for the circuit board and batteries and limit the external wires to only the coil lead and the headphones. Ensure that all external knobs are mounted inside the metal case and that a cover is available to completely enclose the circuit board and controls. Even if you use a plastic case for the PI circuit, you can place the whole thing inside a fully enclosed metal case with the wires exiting in one small hole. Then run a wire from the metal case to a good earth ground near your workbench. That should pretty much eliminate most noise. If you use an external power supply, wrap the DC leads around a ferrite core for a few wraps to attenuate EMI/RFI. Watch the polarity of the supply on the Hammerhead as ground is positive. You may want to ensure that the external power supply floats above ground potential to minimize any problems when a PI control box metal case is grounded. You can also ground the metal case through a capacitor pair in parallel (100 uF in parallel with a .01uF) as an alternative.

2. Make a coil probe which acts like a very small coil. Smaller coils will pick up less broadband EMI/RFI. You could use Eric Foster's recommendations to make a coil probe using a .5" X 3" soft ferrite rod as a core. Ground the ferrite. You can still tweak the circuit on your work bench but all the distances will be reduced to a few inches for small targets. You can orient the probe for the least amount of external EMI/RFI signal pickup around your workbench and have most of the benefits of a Faraday cage when combined with #1 above.

My workbench is near the heater with wires and pipes all around my workbench. I found this setup very handy to tweak PI circuit designs and make circuit adjustments with minimal external EMI/RFI pickup. Large coils are impossible to use near my workbench. The acid test is to take the PI machine out to the field. Field use requires that more practical design considerations be made about case design, weight and whether you want a fully self-contained PI control box or want an external battery pack mounted on your body to lighten the load.

Good luck with your Digital Hammerhead.

bbsailor

[oldkidnewtoy]

High Charles

Hope you don't mind if I tag onto your post as I'm about to embark on a similar project myself with limited EE experience.

I am planning to use an ADC too measure the results, is there a recommended / realistic resolution 8,10 or 12 bit assuming the output from OPAMP is 5v rail to rail.

Can anybody give a few example measurements for objects and distances (still trying for an oscilloscope on eBay)?

Thank Graeme

[Carl-NC]

Extra voltage won't buy you much. Transmit energy is determined by the DC current through the coil, and the speed at which it's turned off. You can improve either of these without raising the battery voltage.

bbsailor's recommendations for noise suppression are excellent.

- Carl