Posted 22 October, 2016

In my last post on this subject, I discussed a modification to the OSU STEM Outreach’s Pulse Detector schematic to eliminate the dual 9V battery supply in favor of a single one, taking advantage of the LM358 op-amp’s single supply operation capability.  However, before actually recommending that the new arrangement be adopted, I wanted to make sure that it would work properly with the OXO ‘Soft-Clip’ finger clip module instead of the one I created for testing in my lab.  It  should work, but as a long-time engineer I have been bitten more than once by the difference between  should and  would! ;-).

So, Prof Anderson was gracious enough to bring a spare OXO clip to our next Outreach session, and since then I have had the opportunity to test this clip with my new circuit, as shown in the following photo.

After fiddling with the tension screw a bit, I got the Soft-Clip tension set properly for my finger, and lo-and-behold, the pulse detector circuit worked out quite nicely, as shown in the following short video clip (as you watch the clip, note that the IR beam from the IR LED is visible as a blue-white glow).

At this time I also made one other minor change.  In the original OSU circuit, the DC blocking capacitor (C2 in the schematic) was a 0.1uF, but in my circuit this value was changed to 0.01uF, because that’s all I had on hand.  In the meantime, however, I got some 0.1uF’s from Mouser, and so my final circuit as shown above incorporates a 0.1uF vice the original 0.01.

The next step in this project is to transfer the detector circuit from plugboard to a more permanent version on perfboard.  This will allow me to demonstrate the new circuit to Prof Anderson and the rest of the OSU STEM Outreach team, to lend credence to the idea of modifying this project’s documentation to eliminate the now-unneeded second supply.  Stay tuned!

Posted 10/24/2016

So, tonight I had the time to finish transferring the pulse detector circuit from my old trusty plugboard to a more permanent medium – i.e. perfboard.  The idea here is to provide the OSU STEM Outreach team with a working pulse detector circuit running from a single 9V battery, as a working example of my recommended modifications to the  circuit being used presently.  The image below shows the perfboard arrangement, and there is also a short video clip of the new pulse detector circuit in action.

The astute observer might notice there are only three wires going from the OXO ‘Soft-Clip’ assembly to the pulse detector circuit.  While I was wiring up the perfboard version, I realized I could eliminate one wire by simply moving the 100-Ohm IR LED current-limiting resistor from the anode (positive) side of the diode to the cathode (negative) end. This allowed me to connect the IR LED anode to the PhotoDiode cathode, and both to the +9V lead.  Implementing this change for all of the current stock of OXO clips may be more than the OSU STEM Outreach crew wants to take on, but I thought I would mention it ;-).  The finished schematic (with the transposed 100-Ohm resistor) is shown below

 

The next (and hopefully final) step in all this is to 3D print a small box for the perfboard circuit.  Because I can’t help myself, I have decided to try printing  a transparent box – not a trivial undertaking for 3D printing.  Stay tuned!

Posted 10/25/2016

As mentioned above, 3D printing a transparent box is a non-trivial undertaking, at least with the current hobbyist FDM (Fused Deposition Modeling) materials and techniques.  After a number of iterations, the best I could do without a  lot of post-processing was a semi-transparent (but still pretty neat) container for the perfboard version of the ‘new-and-improved’ OSU Pulse Detector circuit.

As I have noted previously, the current 3D printing technology makes it easier, faster, and cheaper to go through a number of test cases (literally in this ‘case’!) on the way to a final product, rather than trying to design a final product all in one crack.  Each test takes about 15-30 minutes and just a few cents’ worth of material, and quite often the iterative process illuminates a design problem or opportunity that wasn’t obvious (or even considered) at the start.  In this case, it became apparent after about the 4th iteration that the perfboard should be mounted to the ‘lid’ via printed-on standoffs rather than to the ‘bottom’.

The TinkerCad model for the box and the lid is shown below, as well as the finished product and some of the precursor test boxes.  And, as usual, a short video of the final product.

Frank

 

 

 

Posted 10/16/2016

A week or so ago I participated in an OSU STEM Outreach program that showed high-school students how to build a working pulse detector circuit, using a commonly available op-amp and an IR LED/Photodetector pair.  After an initial presentation, the students were given step-by-step instructions for building the circuit on a small plugboard, and I helped when students ran into trouble.  By the end of the one-hour session, most students were successful, and were able to see the output LED illuminate in time with their pulse – cool!

As I helped out with the class, I was a bit shocked to see that the circuit being build by the students required two 9V batteries wired in series to create a +/-9V supply for the op-amp. At the time I just assumed somebody forgot to spec the op-amp to be one with a common-mode range including ground, and the extra battery was the ‘field-expedient fix’ for the problem. When I asked Prof. Betty Lise Anderson about this, she said that a single supply had been tried at one point, but ‘didn’t work out’ for unspecified reasons.   This piqued my interest, so  decided I would investigate the problem a bit further in my home lab.

As it turned out, the op-amp being used in the students’ circuits was the ubiquitous dual-LM358  in an 8-pin DIP, and this unit does indeed have a common-mode range including ground, so  the single-supply idea should have worked.  Here’s the original OSU circuit (transcribed into DipTrace’s schematic capture format)

And here’s my final single-supply detector circuit

Comparing the two, the only real difference is the addition of the 470K resistor from the inverting input to ground.  In the original circuit, the inverting input was tied directly to ground, while the non-inverting input had a 470K to ground.  This can be a problem, as the input bias current for the LM358 can be on the order of 100 X 10^-9 (100 pico-Amp), which means that the DC voltage at  the non-inverting input due to bias currents could be as much as 50mV or so.  Since this is on the same order of magnitude as the photo-diode signal at this point, there is a real chance the op-amp would never toggle the output.  The dual-supply setup  eliminates  this problem, but at the cost of a second battery.

The final circuit, as laid out on my plug-board, is shown below.

And, because I can, here’s a short movie showing the pulse detector in action ;-).