Improvements – Page 73 – The Smell of Molten Projects in the Morning

DSO150: USB Serial Output

Taking all those pictures of the DSO150 screen reminded me it has a data dump function: press the V/Div and ADJ buttons to squirt configuration, measurements, and trace data from the TX pad on the main board, just in front of the red-black power wires hot-melt glued in place:

DSO150 USB serial adapter - interior — DSO150 USB serial adapter – interior

The picture shows the “before” stage, while I was figuring out where to carve another hole in the case.

NB: The 113-15001-111 DSO150 firmware version includes the serial output option, so you won’t need third-party firmware. Similarly, current PCBs bring the serial pins to neatly labeled header pads. You should refer to the JYETech DSO150 / DSO Shell product page for the details.

After all the cuttin’ and filin’ was done, it looked like this:

DSO150 USB serial adapter - exterior — DSO150 USB serial adapter – exterior

The power switch on the back of the case (top of the picture) disconnects the lithium cell from the charge controller board (now tucked behind the battery) to eliminate any trickle current discharge. Charging the battery thus requires turning that switch on and turning the scope off with its own power switch (along its front edge). Capturing trace data requires having both switches on (duh), whereupon the scope’s normal operating current convinces the charge controller that the cell hasn’t reached full charge. Turn the scope off and, most likely, the controller will tell you the cell is fully charged.

An intro blurb squirts from the port at 115200 in good old 8N1 format when you turn the scope on:

DSO Shell
JYE Tech Ltd.
WWW.JYETECH.COM
FW: 113-15001-111

Pressing the V/Div and ADJ buttons dumps the trace data:

VSen,0.5V
Couple,DC
VPos, -2.02V
Timebase,0.2s
HPos,00362
TriggerMode,NORM
TriggerSlope,Rising
TriggerLevel,  2.02V
RecordLength,01024
Vmax,  2.85V
Vmin,  0.24V
Vavr,  0.87V
Vpp,  2.61V
Vrms,  1.03V
Freq, 0.441Hz
Cycl, 2.266s
PW, 0.231s
Duty, 10.2 %
SampleInterval,00008ms
00000,0000000000, 0.8518688
00001,0000000008, 0.5273474
00002,0000000016, 0.5273474
00003,0000000024, 0.5476300
00004,0000000032, 0.5476300
00005,0000000040, 0.5476300
<< snippage >>
01015,0000008120, 0.8113037
01016,0000008128, 0.8315863
01017,0000008136, 0.8315863
01018,0000008144, 0.8315863
01019,0000008152, 0.8315863
01020,0000008160, 0.8315863
01021,0000008168, 0.8315863
01022,0000008176, 0.8518688
01023,0000008184, 0.8518688

It’s all in neatly comma-separated-value format, so you can slam it into a spreadsheet and have your way with it. Utilities also exist to capture the data, extract the values, and send them directly to GNUplot, etc.

Like so:

If I expected to do a lot of that, I’d boldify the traces and embiggen the text, all of which is in the nature of fine tuning.

It’s hard to reproduce the beauty of the DSO150’s display, though:

The DSO150 remains pretty good for being the worst oscilloscope I’m willing to use …

2020-09-11

Step2 Garden Seat: Replacement Seat2

As expected, the plywood seat I put on the Step2 Garden Seat for Mary’s Vassar Farms plot lasted about a year before the wood rotted away around the screws. In the meantime, we’d acquired a stack of SiLite cafeteria trays, so we applied one to the cause of better seating:

Step2 Seat - tray variant — Step2 Seat – tray variant

Various eBay listings value that slab of ~~Bakelite~~ Melamine up to $20, which is far more than Mary paid for the entire stack at a local tag sale. They also call that color “rich brown”, which is certainly better than what immediately came to mind when I saw them.

The stylin’ asymmetric design happened when I realized the squared-off handle end of the cart didn’t demand a rounded-off end of the seat. I cut off the raised tray rim before sketching the rounded outline using the rotted seat as a template; some of the sketch remains over on the right-front corner. A session with Mr Belt Sander put the remaining rim edges flush with the surface, no matter what the picture suggests.

The tray being 2 mm thinner than the plywood, I tried printing the hinges in a different orientation with different built-in support:

Rolling Cart Hinges - solid model - build — Rolling Cart Hinges – solid model – build

The perimeter threads pulled up far too much and, although fiddling with cooling would likely help, I think the original orientation was better:

Rolling Cart Hinges - solid model - bottom — Rolling Cart Hinges – solid model – bottom

Given that the post-apocalypse breakfast will be served on similar trays, the seat should survive for quite a while in the garden. We think the sun will convert the brown surface into a bun warmer; a coat of white paint may be in its future.

The original OpenSCAD code is still out there as a GitHub Gist.

2020-09-10

Discrete LM3909 LED Flasher: Circuit Variations

The basic discrete LM3909 LED Flasher circuit looks like this:

Discrete LM3909 - basic circuit — Discrete LM3909 – basic circuit

The LM3909 IC boosted a single 1.5 V cell enough to fire a red(-ish) LED, even with the cell well under 1 V. I want to blink a blue(-ish) LED from a pair of AA alkaline cells (with the right size & heft to serve as a base for the hairball circuitry), so the voltage ranges from just over 3 V down to maybe 1.5 V. Although the original circuit works, the LED pulse is long enough to put a reverse bias on the timing capacitor; a 470 µF electrolytic cap (positive terminal on the right at node P2-OUT) produces a pulse every few seconds.

A slightly tweaked version of the circuitry puts -400 mV across C1 (green trace) by the end of the pulse:

Discrete LM3909 - basic circuit - 3.0 V simulation — Discrete LM3909 – basic circuit – 3.0 V simulation

The App Note describes the negative feedback loop from the collector of “power transistor” Q3 through Q4 and Q1, closing through the Q2 current mirror. The base-emitter drops of Q4 and Q1 set the trip point where Q1 starts to conduct and the LED turns on.

Q3 is on when the LED is on, with C1 reverse-charging through R1 and the LED. The voltage at the top of R2 rises from the negative voltage at the start of the pulse, carrying the emitter of Q1 along with it. The LED pulse will end when the rising emitter voltage shuts off Q1 and, thus, the Q2 current mirror driving Q3. Because Q3 holds the bottom of R5 close to 0 V, the base of Q4 is at about half the supply voltage, so Q1 remains on until its emitter rises to about 2 forward drops (handwavingly ignoring the R6 + R7 voltage divider) below the supply.

If the LED pulse is longer than required to completely discharge C1, the poor cap gets reverse-biased and suffers indigestion. Aluminum electrolytics can withstand a little reverse bias, but it’s Bad Practice.

When Q3 and the LED are off, C1 forward-charges through (R4 + R5) + R2, with most of the initial voltage across R2, because C1 should start with a little more than 0 V across it. This holds the current mirror off until C1 charges enough to raise the base of Q4 about two forward drops above Q1’s emitter, shove current through Q4 and Q1, turn on the Q2 current mirror, Q3, and light the LED.

Around and around it goes!

The worst case for reverse charge happens at higher supply (a.k.a. battery) voltages and higher LED currents. Reducing the reverse charge time requires more forward drop through Q4 + Q1 to soak up the higher voltage and lower the trip voltage at Q1’s emitter, which suggests putting another forward-biased junction in series.

Putting a diode in Q1’s base lead doesn’t produce much improvement:

Discrete LM3909 - Q1 B diode - 3.0 V — Discrete LM3909 – Q1 B diode – 3.0 V

Perhaps because the 27 µA current at the trip point is so low the diode doesn’t actually have much forward drop; the simulation says 400 mV.

Putting the diode in the emitter runs the current mirror’s 5 mA through it:

Discrete LM3909 - Q1 E diode - 3.0 V — Discrete LM3909 – Q1 E diode – 3.0 V

The overall period remains about 2 s, but the LED pulse = reverse charge time drops by a factor of two and the cap voltage bottoms out at 0 V, so that’s good.

A Darlington transistor provides far more gain to compensate for the reduced base drive:

Discrete LM3909 - Darl Q1 - 3.0 V — Discrete LM3909 – Darl Q1 – 3.0 V

The LED pulse is slightly shorter and its current goes up a smidge, but the cap voltage remains above zero.

A line in the LM3909 App Note mentions that the Q2 current mirror amplifies Q1’s emitter current by a factor of three: “This current will be amplified by about 3 by Q2 and passed to the base of Q3”. An IC current mirror’s designer can scale its output by varying the collector area, but out here in the discrete world we must splice multiple transistors in parallel:

Discrete LM3909 - Darl Q1 3xQ2- 3.0 V — Discrete LM3909 – Darl Q1 3xQ2- 3.0 V

More base drive in Q3 doesn’t buy much, because it’s already pretty well saturated during the pulse, but the current goes up enough to push C1 slightly into reverse charge territory again. As far as I can tell, the factor-of-three gain was required to make up for the relatively poor performance of IC technology around 1970; things have definitely improved since then.

It’s worth mentioning that the actual circuitry (in particular, the LEDs!) will differ from the simulations, so the pretty plots are more along the lines of serving suggestions than actual predictions. Verily, a simulation can’t prove that a circuit will work, but can sometimes help show why it won’t.

All the LTSpice simulation files tucked into a GitHub Gist:

	Version 4
	SHEET 1 1816 760
	WIRE 64 -16 0 -16
	WIRE 112 -16 64 -16
	WIRE 272 -16 112 -16
	WIRE 544 -16 272 -16
	WIRE 1024 -16 544 -16
	WIRE 1200 -16 1024 -16
	WIRE 112 0 112 -16
	WIRE 272 0 272 -16
	WIRE 1024 32 1024 -16
	WIRE 1200 32 1200 -16
	WIRE 0 48 0 -16
	WIRE 544 48 544 -16
	WIRE 944 80 864 80
	WIRE 960 80 944 80
	WIRE 1136 80 1104 80
	WIRE 272 96 272 80
	WIRE 400 96 272 96
	WIRE 480 96 400 96
	WIRE 400 112 400 96
	WIRE 112 128 112 80
	WIRE 272 128 272 96
	WIRE 688 128 640 128
	WIRE 800 128 752 128
	WIRE 944 144 944 80
	WIRE 1024 144 1024 128
	WIRE 1024 144 944 144
	WIRE 1104 144 1104 80
	WIRE 1104 144 1024 144
	WIRE 112 160 112 128
	WIRE 864 192 864 176
	WIRE 912 192 864 192
	WIRE 0 208 0 128
	WIRE 400 208 400 192
	WIRE 640 208 640 128
	WIRE 640 208 400 208
	WIRE 400 224 400 208
	WIRE 112 272 112 224
	WIRE 128 272 112 272
	WIRE 224 272 192 272
	WIRE 272 272 272 208
	WIRE 272 272 224 272
	WIRE 912 272 912 192
	WIRE 400 336 400 304
	WIRE 544 336 544 144
	WIRE 544 336 400 336
	WIRE 112 368 112 272
	WIRE 912 368 912 352
	WIRE 912 368 112 368
	WIRE 272 400 272 272
	WIRE 112 416 112 368
	WIRE 1200 448 1200 128
	WIRE 1200 448 336 448
	WIRE 112 464 112 416
	WIRE 400 480 400 336
	WIRE 0 496 0 288
	WIRE 0 496 -32 496
	WIRE -32 528 -32 496
	WIRE 0 576 0 496
	WIRE 64 576 0 576
	WIRE 112 576 112 544
	WIRE 112 576 64 576
	WIRE 272 576 272 496
	WIRE 272 576 112 576
	WIRE 400 576 400 560
	WIRE 400 576 272 576
	FLAG -32 528 0
	FLAG 112 128 P6-RLIM
	FLAG 112 416 P18-RC
	FLAG 224 272 P2-OUT
	FLAG 64 -16 P5-V+
	FLAG 64 576 P4-V-
	SYMBOL res 96 -16 R0
	SYMATTR InstName R1
	SYMATTR Value 39
	SYMBOL LED 96 160 R0
	SYMATTR InstName D1
	SYMATTR Value LXHL-BW02
	SYMBOL res 96 448 R0
	SYMATTR InstName R2
	SYMATTR Value 9.1k
	SYMBOL cap 192 256 R90
	WINDOW 0 0 32 VBottom 2
	WINDOW 3 32 32 VTop 2
	WINDOW 40 52 32 VTop 2
	SYMATTR InstName C1
	SYMATTR Value 1m
	SYMATTR SpiceLine2 ic=0.5
	SYMBOL res 256 112 R0
	SYMATTR InstName R5
	SYMATTR Value 390
	SYMBOL res 256 -16 R0
	SYMATTR InstName R4
	SYMATTR Value 390
	SYMBOL npn 336 400 M0
	SYMATTR InstName Q3
	SYMATTR Value 2N3904
	SYMBOL res 384 96 R0
	SYMATTR InstName R6
	SYMATTR Value 20k
	SYMBOL res 384 208 R0
	SYMATTR InstName R7
	SYMATTR Value 10k
	SYMBOL res 384 464 R0
	SYMATTR InstName R8
	SYMATTR Value 20k
	SYMBOL npn 480 48 R0
	SYMATTR InstName Q4
	SYMATTR Value 2N3904
	SYMBOL npn 800 80 R0
	SYMATTR InstName Q1
	SYMATTR Value 2N3904
	SYMBOL pnp 960 128 M180
	SYMATTR InstName Q2a
	SYMATTR Value 2N3906
	SYMBOL voltage 0 192 R0
	WINDOW 123 0 0 Left 0
	WINDOW 39 24 116 Left 2
	SYMATTR InstName V1
	SYMATTR Value 1.5
	SYMBOL res 896 256 R0
	SYMATTR InstName R9
	SYMATTR Value 100
	SYMBOL pnp 1136 128 M180
	SYMATTR InstName Q2b
	SYMATTR Value 2N3906
	SYMBOL res -16 32 R0
	SYMATTR InstName PTC
	SYMATTR Value 5
	SYMBOL diode 688 144 R270
	WINDOW 0 32 32 VTop 2
	WINDOW 3 0 32 VBottom 2
	SYMATTR InstName D2
	SYMATTR Value 1N4148
	TEXT -56 192 VRight 2 ;AA Alkaline
	TEXT 512 528 Left 2 !.tran 30
	TEXT 504 496 Left 2 ;Pins 3 and 7 = no connect
	TEXT 760 488 Left 2 ;Sorted by Vd at If=20 mA \n \nPart # Mfg Is (A) N Iave (A) Vf@Iave (V) Vd@If (V)\nQTLP690C Fairchild 1.00E-22 1.500 0.16 1.90 1.82\nPT-121-B Luminous 4.35E-07 8.370 20.00 3.84 2.34\nLUW-W5AP OSRAM 6.57E-08 7.267 2.00 3.26 2.39\nLXHL-BW02 Lumileds 4.50E-20 2.600 0.40 2.95 2.75\nW5AP-LZMZ-5K Lumileds 3.50E-17 3.120 2.00 3.13 2.76\nLXK2-PW14 Lumileds 3.50E-17 3.120 1.60 3.11 2.76\nAOT-2015 AOT 5.96E-10 6.222 0.18 3.16 2.80\nNSSW008CT-P Nichia 2.30E-16 3.430 0.04 2.92 2.86\nNSSWS108T Nichia 1.13E-18 3.020 0.04 2.99 2.94\nNSPW500BS Nichia 2.70E-10 6.790 0.03 3.27 3.20\nNSCW100 Nichia 1.69E-08 9.626 0.03 3.60 3.50
	TEXT 168 256 Left 4 ;+
	TEXT 656 56 Left 2 ;3 V battery needs more VBE

view raw Discrete LM3909 – D2 Q1.asc hosted with ❤ by GitHub

	Version 4
	SHEET 1 1816 760
	WIRE 64 -16 0 -16
	WIRE 112 -16 64 -16
	WIRE 272 -16 112 -16
	WIRE 544 -16 272 -16
	WIRE 1024 -16 544 -16
	WIRE 1200 -16 1024 -16
	WIRE 112 0 112 -16
	WIRE 272 0 272 -16
	WIRE 1024 32 1024 -16
	WIRE 1200 32 1200 -16
	WIRE 0 48 0 -16
	WIRE 544 48 544 -16
	WIRE 848 80 736 80
	WIRE 944 80 848 80
	WIRE 960 80 944 80
	WIRE 1136 80 1104 80
	WIRE 272 96 272 80
	WIRE 400 96 272 96
	WIRE 480 96 400 96
	WIRE 400 112 400 96
	WIRE 112 128 112 80
	WIRE 272 128 272 96
	WIRE 672 128 640 128
	WIRE 848 144 848 80
	WIRE 944 144 944 80
	WIRE 1024 144 1024 128
	WIRE 1024 144 944 144
	WIRE 1104 144 1104 80
	WIRE 1104 144 1024 144
	WIRE 112 160 112 128
	WIRE 736 192 736 176
	WIRE 784 192 736 192
	WIRE 0 208 0 128
	WIRE 400 208 400 192
	WIRE 640 208 640 128
	WIRE 640 208 400 208
	WIRE 400 224 400 208
	WIRE 912 240 848 240
	WIRE 112 272 112 224
	WIRE 128 272 112 272
	WIRE 224 272 192 272
	WIRE 272 272 272 208
	WIRE 272 272 224 272
	WIRE 912 272 912 240
	WIRE 400 336 400 304
	WIRE 544 336 544 144
	WIRE 544 336 400 336
	WIRE 112 368 112 272
	WIRE 912 368 912 352
	WIRE 912 368 112 368
	WIRE 272 400 272 272
	WIRE 112 416 112 368
	WIRE 1200 448 1200 128
	WIRE 1200 448 336 448
	WIRE 112 464 112 416
	WIRE 400 480 400 336
	WIRE 0 496 0 288
	WIRE 0 496 -32 496
	WIRE -32 528 -32 496
	WIRE 0 576 0 496
	WIRE 64 576 0 576
	WIRE 112 576 112 544
	WIRE 112 576 64 576
	WIRE 272 576 272 496
	WIRE 272 576 112 576
	WIRE 400 576 400 560
	WIRE 400 576 272 576
	FLAG -32 528 0
	FLAG 112 128 P6-RLIM
	FLAG 112 416 P18-RC
	FLAG 224 272 P2-OUT
	FLAG 64 -16 P5-V+
	FLAG 64 576 P4-V-
	SYMBOL res 96 -16 R0
	SYMATTR InstName R1
	SYMATTR Value 39
	SYMBOL LED 96 160 R0
	SYMATTR InstName D1
	SYMATTR Value NSSW008CT-P1
	SYMBOL res 96 448 R0
	SYMATTR InstName R2
	SYMATTR Value 9.1k
	SYMBOL cap 192 256 R90
	WINDOW 0 0 32 VBottom 2
	WINDOW 3 32 32 VTop 2
	WINDOW 40 52 32 VTop 2
	SYMATTR InstName C1
	SYMATTR Value 470�
	SYMATTR SpiceLine2 ic=0.5
	SYMBOL res 256 112 R0
	SYMATTR InstName R5
	SYMATTR Value 390
	SYMBOL res 256 -16 R0
	SYMATTR InstName R4
	SYMATTR Value 390
	SYMBOL npn 336 400 M0
	SYMATTR InstName Q3
	SYMATTR Value 2N3904
	SYMBOL res 384 96 R0
	SYMATTR InstName R6
	SYMATTR Value 20k
	SYMBOL res 384 208 R0
	SYMATTR InstName R7
	SYMATTR Value 10k
	SYMBOL res 384 464 R0
	SYMATTR InstName R8
	SYMATTR Value 20k
	SYMBOL npn 480 48 R0
	SYMATTR InstName Q4
	SYMATTR Value 2N3904
	SYMBOL npn 672 80 R0
	SYMATTR InstName Q1a
	SYMATTR Value 2N3904
	SYMBOL pnp 960 128 M180
	SYMATTR InstName Q2a
	SYMATTR Value 2N3906
	SYMBOL voltage 0 192 R0
	WINDOW 123 0 0 Left 0
	WINDOW 39 24 116 Left 2
	SYMATTR InstName V1
	SYMATTR Value 1.5
	SYMBOL res 896 256 R0
	SYMATTR InstName R9
	SYMATTR Value 100
	SYMBOL pnp 1136 128 M180
	SYMATTR InstName Q2b
	SYMATTR Value 2N3906
	SYMBOL npn 784 144 R0
	SYMATTR InstName Q1b
	SYMATTR Value 2N3904
	SYMBOL res -16 32 R0
	SYMATTR InstName PTC
	SYMATTR Value 5
	TEXT -56 192 VRight 2 ;AA Alkaline
	TEXT 512 528 Left 2 !.tran 30
	TEXT 504 496 Left 2 ;Pins 3 and 7 = no connect
	TEXT 760 488 Left 2 ;Sorted by Vd at If=20 mA \n \nPart # Mfg Is (A) N Iave (A) Vf@Iave (V) Vd@If (V)\nQTLP690C Fairchild 1.00E-22 1.500 0.16 1.90 1.82\nPT-121-B Luminous 4.35E-07 8.370 20.00 3.84 2.34\nLUW-W5AP OSRAM 6.57E-08 7.267 2.00 3.26 2.39\nLXHL-BW02 Lumileds 4.50E-20 2.600 0.40 2.95 2.75\nW5AP-LZMZ-5K Lumileds 3.50E-17 3.120 2.00 3.13 2.76\nLXK2-PW14 Lumileds 3.50E-17 3.120 1.60 3.11 2.76\nAOT-2015 AOT 5.96E-10 6.222 0.18 3.16 2.80\nNSSW008CT-P Nichia 2.30E-16 3.430 0.04 2.92 2.86\nNSSWS108T Nichia 1.13E-18 3.020 0.04 2.99 2.94\nNSPW500BS Nichia 2.70E-10 6.790 0.03 3.27 3.20\nNSCW100 Nichia 1.69E-08 9.626 0.03 3.60 3.50
	TEXT 168 256 Left 4 ;+
	TEXT 664 256 Left 2 ;Use MPSA14 Darlington
	TEXT 656 56 Left 2 ;3 V battery needs more VBE

view raw Discrete LM3909 – Darl Q1 – 1X Q2 .asc hosted with ❤ by GitHub

	Version 4
	SHEET 1 1816 760
	WIRE 64 -16 0 -16
	WIRE 112 -16 64 -16
	WIRE 272 -16 112 -16
	WIRE 544 -16 272 -16
	WIRE 1024 -16 544 -16
	WIRE 1200 -16 1024 -16
	WIRE 1360 -16 1200 -16
	WIRE 1520 -16 1360 -16
	WIRE 112 0 112 -16
	WIRE 272 0 272 -16
	WIRE 1024 32 1024 -16
	WIRE 1200 32 1200 -16
	WIRE 1360 32 1360 -16
	WIRE 1520 32 1520 -16
	WIRE 0 48 0 -16
	WIRE 544 48 544 -16
	WIRE 848 80 736 80
	WIRE 944 80 848 80
	WIRE 960 80 944 80
	WIRE 1136 80 1104 80
	WIRE 1296 80 1280 80
	WIRE 1456 80 1440 80
	WIRE 272 96 272 80
	WIRE 400 96 272 96
	WIRE 480 96 400 96
	WIRE 400 112 400 96
	WIRE 112 128 112 80
	WIRE 272 128 272 96
	WIRE 672 128 640 128
	WIRE 848 144 848 80
	WIRE 944 144 944 80
	WIRE 1024 144 1024 128
	WIRE 1024 144 944 144
	WIRE 1104 144 1104 80
	WIRE 1104 144 1024 144
	WIRE 1280 144 1280 80
	WIRE 1280 144 1104 144
	WIRE 1440 144 1440 80
	WIRE 1440 144 1280 144
	WIRE 112 160 112 128
	WIRE 736 192 736 176
	WIRE 784 192 736 192
	WIRE 1200 192 1200 128
	WIRE 1360 192 1360 128
	WIRE 1360 192 1200 192
	WIRE 1520 192 1520 128
	WIRE 1520 192 1360 192
	WIRE 0 208 0 128
	WIRE 400 208 400 192
	WIRE 640 208 640 128
	WIRE 640 208 400 208
	WIRE 400 224 400 208
	WIRE 912 240 848 240
	WIRE 112 272 112 224
	WIRE 128 272 112 272
	WIRE 224 272 192 272
	WIRE 272 272 272 208
	WIRE 272 272 224 272
	WIRE 912 272 912 240
	WIRE 400 336 400 304
	WIRE 544 336 544 144
	WIRE 544 336 400 336
	WIRE 112 368 112 272
	WIRE 912 368 912 352
	WIRE 912 368 112 368
	WIRE 272 400 272 272
	WIRE 112 416 112 368
	WIRE 1200 448 1200 192
	WIRE 1200 448 336 448
	WIRE 112 464 112 416
	WIRE 400 480 400 336
	WIRE 0 496 0 288
	WIRE 0 496 -32 496
	WIRE -32 528 -32 496
	WIRE 0 576 0 496
	WIRE 64 576 0 576
	WIRE 112 576 112 544
	WIRE 112 576 64 576
	WIRE 272 576 272 496
	WIRE 272 576 112 576
	WIRE 400 576 400 560
	WIRE 400 576 272 576
	FLAG -32 528 0
	FLAG 112 128 P6-RLIM
	FLAG 112 416 P18-RC
	FLAG 224 272 P2-OUT
	FLAG 64 -16 P5-V+
	FLAG 64 576 P4-V-
	SYMBOL res 96 -16 R0
	SYMATTR InstName R1
	SYMATTR Value 39
	SYMBOL LED 96 160 R0
	SYMATTR InstName D1
	SYMATTR Value NSSW008CT-P1
	SYMBOL res 96 448 R0
	SYMATTR InstName R2
	SYMATTR Value 9.1k
	SYMBOL cap 192 256 R90
	WINDOW 0 0 32 VBottom 2
	WINDOW 3 32 32 VTop 2
	WINDOW 40 52 32 VTop 2
	SYMATTR InstName C1
	SYMATTR Value 470�
	SYMATTR SpiceLine2 ic=0.5
	SYMBOL res 256 112 R0
	SYMATTR InstName R5
	SYMATTR Value 390
	SYMBOL res 256 -16 R0
	SYMATTR InstName R4
	SYMATTR Value 390
	SYMBOL npn 336 400 M0
	SYMATTR InstName Q3
	SYMATTR Value 2N3904
	SYMBOL res 384 96 R0
	SYMATTR InstName R6
	SYMATTR Value 20k
	SYMBOL res 384 208 R0
	SYMATTR InstName R7
	SYMATTR Value 10k
	SYMBOL res 384 464 R0
	SYMATTR InstName R8
	SYMATTR Value 20k
	SYMBOL npn 480 48 R0
	SYMATTR InstName Q4
	SYMATTR Value 2N3904
	SYMBOL npn 672 80 R0
	SYMATTR InstName Q1a
	SYMATTR Value 2N3904
	SYMBOL pnp 960 128 M180
	SYMATTR InstName Q2a
	SYMATTR Value 2N3906
	SYMBOL voltage 0 192 R0
	WINDOW 123 0 0 Left 0
	WINDOW 39 24 116 Left 2
	SYMATTR InstName V1
	SYMATTR Value 1.5
	SYMBOL res 896 256 R0
	SYMATTR InstName R9
	SYMATTR Value 100
	SYMBOL pnp 1136 128 M180
	SYMATTR InstName Q2b
	SYMATTR Value 2N3906
	SYMBOL pnp 1296 128 M180
	SYMATTR InstName Q2c
	SYMATTR Value 2N3906
	SYMBOL pnp 1456 128 M180
	SYMATTR InstName Q2d
	SYMATTR Value 2N3906
	SYMBOL npn 784 144 R0
	SYMATTR InstName Q1b
	SYMATTR Value 2N3904
	SYMBOL res -16 32 R0
	SYMATTR InstName PTC
	SYMATTR Value 5
	TEXT -56 192 VRight 2 ;AA Alkaline
	TEXT 512 528 Left 2 !.tran 15
	TEXT 504 496 Left 2 ;Pins 3 and 7 = no connect
	TEXT 760 488 Left 2 ;Sorted by Vd at If=20 mA \n \nPart # Mfg Is (A) N Iave (A) Vf@Iave (V) Vd@If (V)\nQTLP690C Fairchild 1.00E-22 1.500 0.16 1.90 1.82\nPT-121-B Luminous 4.35E-07 8.370 20.00 3.84 2.34\nLUW-W5AP OSRAM 6.57E-08 7.267 2.00 3.26 2.39\nLXHL-BW02 Lumileds 4.50E-20 2.600 0.40 2.95 2.75\nW5AP-LZMZ-5K Lumileds 3.50E-17 3.120 2.00 3.13 2.76\nLXK2-PW14 Lumileds 3.50E-17 3.120 1.60 3.11 2.76\nAOT-2015 AOT 5.96E-10 6.222 0.18 3.16 2.80\nNSSW008CT-P Nichia 2.30E-16 3.430 0.04 2.92 2.86\nNSSWS108T Nichia 1.13E-18 3.020 0.04 2.99 2.94\nNSPW500BS Nichia 2.70E-10 6.790 0.03 3.27 3.20\nNSCW100 Nichia 1.69E-08 9.626 0.03 3.60 3.50
	TEXT 168 256 Left 4 ;+
	TEXT 1224 224 Left 2 ;Current mirror with 3X current gain
	TEXT 664 256 Left 2 ;Use MPSA14 Darlington
	TEXT 656 56 Left 2 ;3 V battery needs more VBE

view raw Discrete LM3909 – Darl Q1 – 3X Q2 .asc hosted with ❤ by GitHub

	Version 4
	SHEET 1 1816 760
	WIRE 64 -16 0 -16
	WIRE 112 -16 64 -16
	WIRE 272 -16 112 -16
	WIRE 544 -16 272 -16
	WIRE 1024 -16 544 -16
	WIRE 1200 -16 1024 -16
	WIRE 112 0 112 -16
	WIRE 272 0 272 -16
	WIRE 1024 32 1024 -16
	WIRE 1200 32 1200 -16
	WIRE 0 48 0 -16
	WIRE 544 48 544 -16
	WIRE 944 80 768 80
	WIRE 960 80 944 80
	WIRE 1136 80 1104 80
	WIRE 272 96 272 80
	WIRE 400 96 272 96
	WIRE 480 96 400 96
	WIRE 400 112 400 96
	WIRE 112 128 112 80
	WIRE 272 128 272 96
	WIRE 704 128 640 128
	WIRE 944 144 944 80
	WIRE 1024 144 1024 128
	WIRE 1024 144 944 144
	WIRE 1104 144 1104 80
	WIRE 1104 144 1024 144
	WIRE 112 160 112 128
	WIRE 768 192 768 176
	WIRE 800 192 768 192
	WIRE 912 192 864 192
	WIRE 0 208 0 128
	WIRE 400 208 400 192
	WIRE 640 208 640 128
	WIRE 640 208 400 208
	WIRE 400 224 400 208
	WIRE 112 272 112 224
	WIRE 128 272 112 272
	WIRE 224 272 192 272
	WIRE 272 272 272 208
	WIRE 272 272 224 272
	WIRE 912 272 912 192
	WIRE 400 336 400 304
	WIRE 544 336 544 144
	WIRE 544 336 400 336
	WIRE 112 368 112 272
	WIRE 912 368 912 352
	WIRE 912 368 112 368
	WIRE 272 400 272 272
	WIRE 112 416 112 368
	WIRE 1200 448 1200 128
	WIRE 1200 448 336 448
	WIRE 112 464 112 416
	WIRE 400 480 400 336
	WIRE 0 496 0 288
	WIRE 0 496 -32 496
	WIRE -32 528 -32 496
	WIRE 0 576 0 496
	WIRE 64 576 0 576
	WIRE 112 576 112 544
	WIRE 112 576 64 576
	WIRE 272 576 272 496
	WIRE 272 576 112 576
	WIRE 400 576 400 560
	WIRE 400 576 272 576
	FLAG -32 528 0
	FLAG 112 128 P6-RLIM
	FLAG 112 416 P18-RC
	FLAG 224 272 P2-OUT
	FLAG 64 -16 P5-V+
	FLAG 64 576 P4-V-
	SYMBOL res 96 -16 R0
	SYMATTR InstName R1
	SYMATTR Value 39
	SYMBOL LED 96 160 R0
	SYMATTR InstName D1
	SYMATTR Value LXHL-BW02
	SYMBOL res 96 448 R0
	SYMATTR InstName R2
	SYMATTR Value 9.1k
	SYMBOL cap 192 256 R90
	WINDOW 0 0 32 VBottom 2
	WINDOW 3 32 32 VTop 2
	WINDOW 40 52 32 VTop 2
	SYMATTR InstName C1
	SYMATTR Value 1m
	SYMATTR SpiceLine2 ic=0.5
	SYMBOL res 256 112 R0
	SYMATTR InstName R5
	SYMATTR Value 390
	SYMBOL res 256 -16 R0
	SYMATTR InstName R4
	SYMATTR Value 390
	SYMBOL npn 336 400 M0
	SYMATTR InstName Q3
	SYMATTR Value 2N3904
	SYMBOL res 384 96 R0
	SYMATTR InstName R6
	SYMATTR Value 20k
	SYMBOL res 384 208 R0
	SYMATTR InstName R7
	SYMATTR Value 10k
	SYMBOL res 384 464 R0
	SYMATTR InstName R8
	SYMATTR Value 20k
	SYMBOL npn 480 48 R0
	SYMATTR InstName Q4
	SYMATTR Value 2N3904
	SYMBOL npn 704 80 R0
	SYMATTR InstName Q1
	SYMATTR Value 2N3904
	SYMBOL pnp 960 128 M180
	SYMATTR InstName Q2a
	SYMATTR Value 2N3906
	SYMBOL voltage 0 192 R0
	WINDOW 123 0 0 Left 0
	WINDOW 39 24 116 Left 2
	SYMATTR InstName V1
	SYMATTR Value 1.5
	SYMBOL res 896 256 R0
	SYMATTR InstName R9
	SYMATTR Value 100
	SYMBOL pnp 1136 128 M180
	SYMATTR InstName Q2b
	SYMATTR Value 2N3906
	SYMBOL res -16 32 R0
	SYMATTR InstName PTC
	SYMATTR Value 5
	SYMBOL diode 800 208 R270
	WINDOW 0 32 32 VTop 2
	WINDOW 3 0 32 VBottom 2
	SYMATTR InstName D2
	SYMATTR Value 1N4148
	TEXT -56 192 VRight 2 ;AA Alkaline
	TEXT 512 528 Left 2 !.tran 30
	TEXT 504 496 Left 2 ;Pins 3 and 7 = no connect
	TEXT 760 488 Left 2 ;Sorted by Vd at If=20 mA \n \nPart # Mfg Is (A) N Iave (A) Vf@Iave (V) Vd@If (V)\nQTLP690C Fairchild 1.00E-22 1.500 0.16 1.90 1.82\nPT-121-B Luminous 4.35E-07 8.370 20.00 3.84 2.34\nLUW-W5AP OSRAM 6.57E-08 7.267 2.00 3.26 2.39\nLXHL-BW02 Lumileds 4.50E-20 2.600 0.40 2.95 2.75\nW5AP-LZMZ-5K Lumileds 3.50E-17 3.120 2.00 3.13 2.76\nLXK2-PW14 Lumileds 3.50E-17 3.120 1.60 3.11 2.76\nAOT-2015 AOT 5.96E-10 6.222 0.18 3.16 2.80\nNSSW008CT-P Nichia 2.30E-16 3.430 0.04 2.92 2.86\nNSSWS108T Nichia 1.13E-18 3.020 0.04 2.99 2.94\nNSPW500BS Nichia 2.70E-10 6.790 0.03 3.27 3.20\nNSCW100 Nichia 1.69E-08 9.626 0.03 3.60 3.50
	TEXT 168 256 Left 4 ;+
	TEXT 656 56 Left 2 ;3 V battery needs more VBE

view raw Discrete LM3909 – Q1 D2.asc hosted with ❤ by GitHub

Version 4 SHEET 1 1812 680 WIRE 96 -16 32 -16 WIRE 144 -16 96 -16 WIRE 304 -16 144 -16 WIRE 544 -16 304 -16 WIRE 752 -16 544 -16 WIRE 144 0 144 -16 WIRE 304 0 304 -16 WIRE 752 0 752 -16 WIRE 928 0 752 0 WIRE 752 32 752 0 WIRE 928 32 928 0 WIRE 544 48 544 -16 WIRE 832 80 816 80 WIRE 864 80 832 80 WIRE 304 96 304 80 WIRE 400 96 304 96 WIRE 480 96 400 96 WIRE 400 112 400 96 WIRE 144 128 144 80 WIRE 304 128 304 96 WIRE 752 144 752 128 WIRE 752 144 704 144 WIRE 832 144 832 80 WIRE 832 144 752 144 WIRE 144 160 144 128 WIRE 704 160 704 144 WIRE 32 208 32 -16 WIRE 400 208 400 192 WIRE 640 208 400 208 WIRE 400 224 400 208 WIRE 704 272 704 256 WIRE 144 288 144 224 WIRE 160 288 144 288 WIRE 256 288 224 288 WIRE 304 288 304 208 WIRE 304 288 256 288 WIRE 400 320 400 304 WIRE 544 320 544 144 WIRE 544 320 400 320 WIRE 144 368 144 288 WIRE 704 368 704 352 WIRE 704 368 144 368 WIRE 304 400 304 288 WIRE 144 416 144 368 WIRE 928 448 928 128 WIRE 928 448 368 448 WIRE 144 464 144 416 WIRE 400 480 400 320 WIRE 32 496 32 288 WIRE 32 496 0 496 WIRE 0 528 0 496 WIRE 32 576 32 496 WIRE 96 576 32 576 WIRE 144 576 144 544 WIRE 144 576 96 576 WIRE 304 576 304 496 WIRE 304 576 144 576 WIRE 400 576 400 560 WIRE 400 576 304 576 FLAG 0 528 0 FLAG 144 128 P6-RLIM FLAG 144 416 P18-RC FLAG 256 288 P2-OUT FLAG 96 -16 P5-V+ FLAG 96 576 P4-V- SYMBOL res 128 -16 R0 SYMATTR InstName R1 SYMATTR Value 12 SYMBOL LED 128 160 R0 SYMATTR InstName D1 SYMATTR Value PT-121-B SYMBOL res 128 448 R0 SYMATTR InstName R2 SYMATTR Value 9.1k SYMBOL cap 224 272 R90 WINDOW 0 0 32 VBottom 2 WINDOW 3 32 32 VTop 2 WINDOW 40 52 32 VTop 2 SYMATTR InstName C1 SYMATTR Value 470� SYMATTR SpiceLine2 ic=0.5 SYMBOL res 288 112 R0 SYMATTR InstName R5 SYMATTR Value 390 SYMBOL res 288 -16 R0 SYMATTR InstName R4 SYMATTR Value 390 SYMBOL npn 368 400 M0 SYMATTR InstName Q3 SYMATTR Value 2N3904 SYMBOL res 384 96 R0 SYMATTR InstName R6 SYMATTR Value 20k SYMBOL res 384 208 R0 SYMATTR InstName R7 SYMATTR Value 10k SYMBOL res 384 464 R0 SYMATTR InstName R8 SYMATTR Value 20k SYMBOL npn 480 48 R0 SYMATTR InstName Q4 SYMATTR Value 2N3904 SYMBOL npn 640 160 R0 SYMATTR InstName Q1 SYMATTR Value 2N3904 SYMBOL pnp 816 128 R180 SYMATTR InstName Q2a SYMATTR Value 2N3906 SYMBOL voltage 32 192 R0 WINDOW 123 0 0 Left 0 WINDOW 39 24 116 Left 2 SYMATTR InstName V1 SYMATTR Value 3 SYMBOL res 688 256 R0 SYMATTR InstName R9 SYMATTR Value 100 SYMBOL pnp 864 128 M180 SYMATTR InstName Q2b SYMATTR Value 2N3906 TEXT -24 192 VRight 2 ;AA Alkaline TEXT 512 528 Left 2 !.tran 10 TEXT 504 496 Left 2 ;Pins 3 and 7 = no connect TEXT 1064 112 Left 2 ;Sorted by Vd at If=20 mA \n \nPart # Mfg Is (A) N Iave (A) Vf@Iave (V) Vd@If (V)\nQTLP690C Fairchild 1.00E-22 1.500 0.16 1.90 1.82\nPT-121-B Luminous 4.35E-07 8.370 20.00 3.84 2.34\nLUW-W5AP OSRAM 6.57E-08 7.267 2.00 3.26 2.39\nLXHL-BW02 Lumileds 4.50E-20 2.600 0.40 2.95 2.75\nW5AP-LZMZ-5K Lumileds 3.50E-17 3.120 2.00 3.13 2.76\nLXK2-PW14 Lumileds 3.50E-17 3.120 1.60 3.11 2.76\nAOT-2015 AOT 5.96E-10 6.222 0.18 3.16 2.80\nNSSW008CT-P Nichia 2.30E-16 3.430 0.04 2.92 2.86\nNSSWS108T Nichia 1.13E-18 3.020 0.04 2.99 2.94\nNSPW500BS Nichia 2.70E-10 6.790 0.03 3.27 3.20\nNSCW100 Nichia 1.69E-08 9.626 0.03 3.60 3.50

view raw Discrete LM3909.asc hosted with ❤ by GitHub

2020-09-07

NPN RGB Astable Multivibrator: End-of-Life Battery Runtime

After un-wedging the astable’s synchronized periods and recharging the harvested 18650 lithium cell, I put a sock (not shown) over the radome and let it blink on a corner of the desk:

Astable - 10 11 12 uF tweak - 027 — Astable – 10 11 12 uF tweak – 027

The periods are much too short and the NPN astable currents much too high, but the thing runs for about ten days before the over-discharge circuit shuts it down.

So a single NPN astable driving a single-color LED with a more reasonable period should get a month or so from an end-of-life 18650 cell and a MOSFET astable might run for two months.

2020-09-04

Discrete-transistor LM3909 LED Flasher

I’ve been using not-dead-yet lithium batteries to power astable multivibrators blinking LEDs on the red-to-yellow end of the spectrum, because the over-discharge protection circuitry in the batteries shuts down at 2.5 V, while not eking much light from LEDs toward the blue end of the spectrum.

Back in the late 60s, when integrated circuits were new, National Semiconductor designed and, in the early 70s, introduced the LM3909: “a monolithic oscillator specifically designed to flash Light Emitting Diodes”. The IC used an electrolytic capacitor as both timing element and voltage booster by charging the cap, then switching it in reverse series with the LED, to produce a voltage drop larger than the 1.5 V battery supply. The original National Semiconductor LM3909 datasheet will get you started and Application Note 154 gives more details and insight.

Rob Paisley’s work from 2008 suggested a discrete-transistor version might look just as attractive, in a techie sort of way, as the astables, and perhaps boost the 2 V from a pair of not-dead-yet alkaline cells high enough to light a blue LED.

Some LTSpice twiddling produces a suitable circuit:

The labeled nodes correspond to pin numbers on the IC package, with a suffix indicating what they did for a living. R2 combines the two timing resistors in the IC into a single unit, so “P18-RC” combines the pins. The Q2 pair over on the right forms a current mirror driving Q3, which the doc calls the “power transistor”, to yank the positive end of the capacitor to ground to light the LED.

The LED is faked by a PT-121-B diode with a 2.34 V forward drop at 20 mA. It’s rated for 20 A average current, so it’s not a particularly good model for a piddly 5 mm LED, but I’ll define it to be Good Enough for now.

Running the simulation at 1.5 V is encouraging:

Discrete LM3909 - basic circuit - 1.5 V simulation — Discrete LM3909 – basic circuit – 1.5 V simulation

The green trace gives the voltage across the capacitor. Under these conditions, the voltage stays positive, although not by much.

Running it from a 3 V supply changes the results:

The cap charges to about the same voltage, but the pulse now lasts long enough to charge it nearly half a volt in the wrong direction. This is Bad Practice, even though my similarly offending astables have been doing it for years.

The data sheet points out that the forward drops of Q1 and Q2 determine the trigger level for the start of the LED pulse, so adding another forward-biased junction in series should let the cap charge to a higher voltage and, for the same pulse duration, pull the low end up above zero to increase overall happiness.

2020-09-03

Lathe-straightened Copper Wire

I formerly straightened the copper wire into “bus bars” for the astable multivibrators by whacking it with a slide hammer, but someone whose name is lost in the mists of time told me the right way to do it:

Lathe-straightening Wire - setup — Lathe-straightening Wire – setup

Yup, grab a piece of wire at both ends in Tiny Lathe and give it a few low-speed turns while pulling firmly on the tailstock.

No muss, no fuss, no drama, just bar-straight and slightly work-hardened copper wires:

Lathe-straightening Wire - results — Lathe-straightening Wire – results

I slide-hammered the top wire before remembering the clue. The bottom two wires have peppermint-stick swirls.

Thank you, whoever you were!

2020-09-02

PTC Fuses

Lithium battery packs have overcurrent protection cutouts, but alkaline cells depend on their internal resistance and may overheat in response to a serious short circuit. So adding a PTC fuse to the circuitry over an alkaline battery case seemed appropriate:

Discrete LM3909 - Darl Q1 - 1X Q2 - blue LED test — Discrete LM3909 – Darl Q1 – 1X Q2 – blue LED test

That’s a test setup for a discrete-transistor version of an LM3909 LED blinker, about which more later. The PTC fuse looks a lot like a ceramic capacitor with one leg caught in an alligator clip.

Two bags of PTC fuses recently arrived from halfway around the planet, rated at 100 mA and 170 mA. One allegedly came from JinKe and the other probably didn’t pass through a Littelfuse factory despite its part number, but the only datasheet I can find is for the Littelfuse RXEF PTC PolySwitch series, which is surely close enough.

I set up a torture test involving a bench power supply and an ammeter, both offscreen and left to your imagination:

At 75 °F:

100 mA PTC – 4.75 Ω
170 mA PTC – 2.80 Ω

With a dead short simulated by 3 V from the supply, the current stabilized at:

100 mA PTC – 125 mA
170 mA PTC – 135 mA

Cranked to 5 V for that good old TTL vibe:

100 mA PTC – 70 mA
170 mA PTC – 85 mA

The datasheet says they’re good up to 60 V, but that’s just crazy talk.

The abuse put a shiny gloss on the epoxy coating, sort of like when you overcooked one of those wax-insulated capacitors back in the day.

Despite that, a PTC fuse is better than a dead short, if only because the plastic battery case won’t get all melty with the batteries supplying less than half a watt.

2020-08-31