Author’s Note: not so much a blog post as it is a list of random thoughts. I know, I know, I could’ve made more pretty equations, but I typed this on a note pad, and now it’s 3:17 in the morning.
Stuff that I should have understood last month:
logic level transistors are transistors that can be turned on (or off) by a Voltage_gate-drain voltage less than or around 5 volts. This allows it to be controlled augmented (or regular) MCUs like arduinos and Rasperry pi without needing to amplify the voltage on the pwm signal.
Pick N-channel and P-channel, either with enhancement modes
Simple equations on making sure your MOSFETs don’t burn up. (otherwise stated: “Do I need to heat-sink this?”)
via Joule heating if
P = i2*Rds , where Rds is the drain source resistance which is based on gate voltage.
And if your mosfet current form source to drain works at.. idk. 6A, and has a Rds = 0.6667 Ohms value than power consumption is
P = 2.4W
Pd = (max(Tj)-TA) / R_tethaJA, this is max power mosfet can dissipate, where Tj is the junction temperature, and Ta is the ambient temperature.
P = 175-25C / 62.6 c/w, 25c is ambient temp.
then Pd = 35mW which is less than 2.4W
so you’re safe to operate without a heatsink
Vgs or Vth
Rds
P=Rds*i^2
Pd = as above
mosfet shouldn’t have a very high ON high resistance
so…
https://www.baldengineer.com/p-channel-mosfet-tutorial-with-only-positive-voltages.html
I don’t need to worry about using an opAmp inverter (which can go up to 5$/USD!) to get inverted voltages to drive a PNP MOSFET.
However, issues exist when the logic voltage is 5V on the MCU if my Li-Po voltages vary from 9 to 11.1 to 12.75 volts. [9V-12.75V] is basically my absolute range in order for things to be safe. Less than 9Vs (or less than 3V per cell) and you’ll have to pay a visit to the trickle charge fairy. (Less than 2.X V and some chargers will refuse to charge them). More than 12.75V (or 4.24V/cell) and your Li-Pos could could get damaged in some other way.
Anyways, if the source is at 11.1V and the gate is at the max possible voltage of 5, then the difference is 6.1V. The PNP will forever remain on high and never close.
According to this
http://electronics-diy.com/electronic_schematic.php?id=1012
I can use use a simple BJT to account for the difference. In this case, I only care about the voltage, not the current. Actually, I doubt that little transistor can handle more than an Amp. According to the secs,
https://www.sparkfun.com/datasheets/Components/2N3904.pdf
that’s a definite no on going over 1A. So the in-series resistor would have to be something that limits the current going through it. Something not in those URLs Is where said resistor could be placed. In theory, I should be able to place a resistor in series to the 2n3904 on the ground side to limit current as low as possible, but also to keep the voltage value at the source to as close to the 11.1V (incoming voltage source value) as possible. I mean, in my head (and I haven’t drawn this out in paper yet), that should work. I don’t want the li-pos to drain current while any particular MOSFET is off, so ideally that resistor value should be high.
Further, using a transistor to control another transistor will add in some extra delay on the PWM side of things. I’m leaning towards using the 2n3904 to simply on/off the PNP MOSFET.
Oh, the application for all of this is a Peltier h-bridge. e.g. an electric-thermal device.
So the input would be digital (1 = HIGH, ditital 0 = LOW) –→ 2n3904 –→ PNP
While on the bottom 2 NPN MOSFETs, the full fury of a the PWM input will be direct driver of them.
I hope this makes sense. If it doesn’t, it’s because I didn’t put up any specs, it’s 2:24 in the AM, and my equations should probably reference specific pages on particular datasheets. Right, if you didn’t understand it, it’s because it’s my fault. That’s what some professors say anyways (“I’m sorry I didn’t explain it well, let me try again”) If I were lecturing, I would be like, pay attention kiddo! I’m only going to go through this once. Blah blah. So, if you were wondering what was on my mind this weekend, it was on this. These were some of the things that I should’ve known last month, but didn’t until recently.
–
Hamilton is coming to town around May. Ticket prices are already at the 800USD+ range now. The good seats are gone, and the remaining ones range between 1.1K-1.2K. Assuming one takes their S.O. to such a play, we’re talking about paying between 1.6k-2.4k… $_$. Here’s the thing about going to see Hamilton. If you take a girl to see it, she better be your girlfriend, or fiancé, or wife, or something that places the ‘significant’ in the S.O. Buy your tickets now, and the lady friend you’re seeing could be very well be seeing some else within 2-3 months. It’s not for the light hearten relationship. If a guy takes you to see Hamilton, that’s a serious date. That’s a, “hey, will you marry me date?”, not a “Hey, will you be my girlfriend date?” So, here I am wondering.. Screw S.O’s. I’ll go see Hamilton myself. Alone. Solo. But who does that? I’ve went to see movies solo before (sorta like a dating yourself type of thing), but that’s only ~15 bucks. Seeing a play is more like a social event between viewers when you’re dropping that much. I guess the only thing left to say would be:
I,
Want to be in the room when it happens,
the room when it happens.
When movie X came out, some of my friends had already seen it. It was really a family event for some of them (inter-generational type movie). A few of them had even went on to see it 1, 2, even 3 times. I eventually went to see movie X with my own family. I guess, I’m bringing this up because, that same afternoon I had a breakup with a person in my life. We had spoken, made plans, canceled, made more plans, on watching movie X. Tickets may or may have not already been purchased by me. I had planned to go watch it twice. Once with family, the other with the aforementioned. So here I am, comparing movie X tickets with Hamilton tickets. As 45 would say; sad.
How does that tie in to today? I’ve tried asking some people if they wanted to watch movie Y. Similar story, many folks have already seen it (my folks already went without me!) at this point. The longer I wait the subset of friends I have who would have seen it and wouldn’t bother seeing it for the n-th time with me decreases. It’s been Z weeks now. I think I’ll just go watch it myself.
Or.. Maybe I’ll go see it next week? This is supposed to be a hell week for me, a 4 hours of sleep every day type of week. Not really though.. I mean, I don’t need to go to the coding challenge thing on Tuesday, and I don’t need to go to the hackathon thing next week. I sorta need to go to the hackathon thing this weekend though since since tickets have already been paid for.
3/5 Monday: Dr.K meet, design stuff, lecture, design stuff, tech event that ends at 9pm (but I only need to spend 20 minutes there if any at all).
3/6 Tuesday: Dr.R meet, design stuff until whenever (cancel? Go watch movie Y?)
3/7 Wednesday: hackathon stuff, lecture, design stuff
3/8 Thursday: design stuff (end night watching movie Y?)
3/9 Friday: hackathon stuff (fuck sleep)
3/10 Saturday: hackathon stuff (no sleep)
3/11 Sunday: hackathon stuff (sleep? Lol), end night with design stuff
3/12 Monday: Dr.K meet (again), lecture, hackathon #2? (I still need to do design stuff), prep presentation
3/13 same as 3/6, prep. presentation
3/14 similar to 3/7 but team needs to present
I’m not obligated to do hackathon#2 (and now I see why hackathon#1 cost money to get in) from the 12th-16th, so I might cancel that event depending on how much progress on my ‘design stuff’ gets done. How ready my ‘design stuff’ team is on our presentation on the 14th will be the determining factor for that. This leaves tuesday or thursday for movie going this or next week. Hopefully movie Y is still screening by then. Off to go check Fandango.