I decided to put some power to my circuit and see what kind of power readings I got if any. I dug up an old Radio Shack multi voltage wall wart. Supposed to be 3 - 12v but puts out 5 - 17.5v and is rated at 800mA. It has a removable tip so you can change polarity and at one time probably had an assortment of tips but I found it at a thrift store years ago sans extra tips.
Not sure how meaningful this data is but at least I know power is getting through so that part works. The No Load readings are straight off the end of the plug.
| Volts | mA | No Load mA |
| 5.2 | 1.0 | 1.2 |
| 7.2 | 1.7 | 2.0 |
| 9.2 | 2.5 | 2.7 |
| 11.2 | 3.3 | 3.5 |
| 13.3 | 4.1 | 4.4 |
| 17.8 | 5.8 | 6.0 |
View attachment 4459061
Anybody know what type of diode these clear ones are? No numbers on them and have what looks like a fine fuse wire inside.
View attachment 4459063
PS: Those diodes in the phone base turned out to be 1n4003s so I got 5 of those but they are no use atm.
Bed time!
Idk, they look like little zeners. I think the tall 1 on the left standing vertical might be a fuse?
I'm pretty sure it's supposed to hold a constant current regardless the input voltage applied (up to 40V I think I saw on data sheet) and regardless the size of the load (within reason).
Couple things to try:
1. Remove diode and large resistor. With 43Ω you should see ~1.5 mA (0.064 ÷ 43Ω = 0.00148A, 64mV comes from data sheet; room temp is assumed ~76°F, or 298°K; 298°K × 217μV/°K = 0.06466V). Try different voltages, see if the different voltage sources from your AC/DC adapter produce different mA measurements. I'm pretty sure it should stay relatively constant.
2. Replace red/clear diode with the 1n4003. Try cct again with different voltages and measure to see if mA changes. The equation for the R values and expected set current will change, but whatever the actual output mA ends up being, it shouldn't change with different voltage input applied, I think that's what I'm comprehending. If the mA output is constant throughout the gambit of different voltages applied, then your original diode is not functioning as intended. If your mA output changes with different voltages applied, well then, idk lol, perhaps you'd need a voltage regulator on the input of your LM334, but that just seems unideal..
The whole purpose of the second larger resistor + diode is to reduce the constant current fluctuation as the temperature changes. The constant current source can be made with only 1 set resistor, and should hold constant as long as the chip doesn't heat up or the ambient change temp much.
The second large resistor can be changed out for bigger or smaller depending on the change in constant current with temperature rise. So if you heated the chip up and your constant current increased then you'd +/- the size of the big resistor (R2) to offset. A zero temperature coefficient means that theres 0 change in characteristics with temperature fluctuation. A positive tempco means as you increase in temp so does your mA or mV, and a negative tempo means as you increase in temp your mA or mV drop, in this cct we're only worried about mA. You'd add resistance, or use a bigger R2 if your mA were increasing with temp rise, and you'd reduce resistance or make your big resistor (R2) smaller if mA were decreasing with temp rise. Once you dial in your R2 size needed for 0 tempco (using a 1kΩ POT for R2 could be an easy way of determining R2 value, you'd measure mA at ambient, then use a hairblowdryer or something and increase temp, then rotate the POT till you measured the same mA as with ambient), then you could calculate your R2:R1 ratio, and from that you should be able to derive your Iset equation..
Iset
=
{[(0.067V)×(R2:R1 ratio)] + (Vf of diode) + (0.067V)
} /
[R1 × (R2:R1 ratio)
]
All of this can be done without the math, and by trial and error. If you know your desired mA or desired constant current needed for your CS factory then you'd just build your cct with the diode you have (1n4003), then use a 100Ω pot for R1, and a 1kΩ POT for R2, then you'd set up your multimeter to measure output current, and then adjust POT R1 till your output matched the desired current you'd like. Then heat up the chip and adjust POT R2 to match the same mA output that was observed with no heat applied. At least that's my comprehension. I don't have any lm334's or I'd be breadboarding and trying to determine for myself.
Papa Smurf every year! lol
