Gas station without pumps

2012 December 28

Parts packed, pressure sensors assembled, more thoughts on power amp

Filed under: Circuits course,Pressure gauge — gasstationwithoutpumps @ 22:54
Tags: , , , , ,

The DigiKey order arrived today, so I spent some time repacking all the parts into DIP tubes and little plastic bags.  I also assembled and soldered all 12 of the breakout boards for the MPX2053DP pressure sensors, but I’ve not tested the assembled boards yet.  I have an instrumentation amp set up for the boards, that I used when I was debugging the instrumentation amp lab and the breakout board with my own MPX2053DP pressure sensor, so I should be able to run through the dozen tomorrow fairly quickly (as long as there are no problems).

I was surprised by one thing in the DigiKey order, though I shouldn’t have been.  Because they were out of the original IR emitter I had chosen, I had quickly substituted the next cheapest one.  I had not noticed that it was a 5mm part instead of a 3mm part.  That doesn’t matter for this course (we’ll be sticking them in a breadboard anyway), and it might actually turn out to be an advantage, since the photodiode looks a lot like the IR emitter, but is a 3mm part.  It might have been difficult to keep them straight if both had been 3mm parts.

Incidentally, the alligator clips that I bought turned out to have even worse quality control than I expected.  The box of 100 only had 98 alligator clips, and only 96 screws, so two of the clips ended up screwless. The plastic red and black sleeves are looser than on other alligator clips I’ve had, and the whole clip seemed a little flimsy.  I still think they were a reasonably good deal for the low price, but they are definitely low-end parts.

Yesterday I drilled and cut 12 electrode spacers for the stainless steel electrodes out of an old plastic cutting board, but I decided to buy 24″ bolt cutters to cut the 1/8″ stainless steel welding rod, rather than using the cutoff wheel on my wife’s Dremel motor tool.  The bolt cutters should arrive by the middle of next week, and I’ll finish making the stainless steel electrodes then.  I think that the only other lab equipment that is needed is a hotpot for boiling water (I’ll donate the old one from my office, if it still works), an ice bucket, some thermoses, and disposable coffee cups for the first lab.  The lab support staff have already put the secondary containment tubs in the lab.

I’ve also arranged for some stock NaCl solutions (1M,0.1M, and 0.01M) for the electrode lab, so I think the equipment situation is pretty good—I’ll have to convince my son to go in one more time to finish setting up the other 6 computers to run Python and the Arduinos correctly.  It’s looking like he’ll have the data logger code done this weekend, so we are tentatively planning to go in on Monday to install it. If our Leonardo Arduino board hasn’t arrived by then we may delay a day or two so that we can test the data logger customization with it—there are so many differences on the Leonardo board that something is almost bound to break in the code.  We should test on an Arduino Mega board as well, but I wasn’t willing to spend that much just for testing, as I don’t expect any of the students in the class to be bringing in a Mega board.

I also spent a fair amount of time this morning playing with the power amp design and learning to use the Bitscope USB oscilloscope better.  The function generator on the Bitscope Pocket Analyzer is not very good: it can produce adequate sine waves into a high-impedance resistive load, but it had trouble with triangle waves and with driving the input to the amplifier (which I thought was pretty high impedance).  I ended up switching back to my Elenco FG500 function generator, which seems to produce better waveforms, at least in this application.  I also noticed that the Bitscope picks up a lot of noise when measuring low-level signals (about 1mV of noise, even on a 10mV signal).  I compared the Bitscope display to the display on my Kikusui COS5060 for the same signal and found that most of the noise was not present in the source.  It was probably being generated either in the Bitscope leads or in the input stage of the Bitscope A to D.

I was able to drive the 8Ω speakers that the students will be using down to about 3 or 4Hz and up to about 240kHz (though, of course, I couldn’t hear anything above 15kHz).  I think that there may still be some crossover distortion in the amp, so I’m going to rewire it tomorrow with a trimpot in the middle of the bias network to see if I can eliminate the crossover problem without overheating the transistors.  The PTN7800W regulator I’m using to provide 6.6v doesn’t start limiting the current until 3.2A, so I could dissipate as much as 32W in the FETs if I’m not careful.  I started them smoking once (they recovered without signs of permanent damage), and I don’t want to do that again.  I’ll have to make sure that the lab writeup explains how to set the current limits on the bench power supplies and that the students limit their current to 0.5A until they are sure their design is working.

One thing I noticed today is that there was a quiet oscillation around 4.4KHz, probably due to the large delay in driving the capacitive load of the FET gates.  I added a compensating capacitor from the op amp output to the negative input, just by trial and error (0.1µF seems ok, though I suspect I could go smaller), since computing the correct compensation is rather messy.  I should probably read Intersil Application note AN9415.3 (or some similar App Note) to learn how to do the compensation correctly, but it is certainly beyond the scope of this course.  I’m now going to have to think very carefully about how the power amp design task is going to be formulated, so that it is doable by the students but not cookbook.

Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: