Did you buy a fake or defective LM317 or 22 ohms resistor from "over there"?

I have another LM317T here. I'll test it when I get home!

And about the resistor, I already tried a few. I actually tried 2 that I salvaged from some board and summed around 18Ω (15Ω + 3Ω) But the 22Ω were purchased in a local shop, iirc.

 

Did you buy a fake or defective LM317 or 22 ohms resistor from "over there"?

Older UV LEDs had emissions in the purple with a tail of emissions into the .u.v. Their forward voltage was in the 3.3v range. Newer LEDs are available with peak emissions down to 254nm and Vf as high as 10v.
If the OP has properly wired his LM317 constant current source, he should be able to crank up the input voltage until he gets a constant 56mA output.

 

Older UV LEDs had emissions in the purple with a tail of emissions into the .u.v. Their forward voltage was in the 3.3v range. Newer LEDs are available with peak emissions down to 254nm and Vf as high as 10v.
If the OP has properly wired his LM317 constant current source, he should be able to crank up the input voltage until he gets a constant 56mA output.

This is the description of the LEDs I bought


Link here (hope I'm allowed):
https://www.aliexpress.com/item/4000076841074.html

I just tried with another LM317 and only from 14V~15V I can get around the 60mA of current into the 3 branches.
So, That's what I'm going to use!

 

You might have the collector and emitter of the transistor reversed then its base current is way too high which reduces the base voltage. The pins of a European BC548 are CBE but the pins of American 2Nxxxx transistors is EBC.

You deduced that from the schematic?

 

deduced that from the schematic?

That's the norm around here you should see us when not provided a schematic.

 

You deduced that from the schematic?

Yup.

The schematic is everything. It tells us everything about the circuit, and many things about the person. That's why people around here (ok, maybe just me) get a bit cranky when someone posts a question like "Why won't my oscillator circuit work?" but *no schematic*.

Paraphrasing Rear Admiral Joshua Painter,

"Engineers don't take a dump, son, without a schematic."

ak

 

I asked that because the schematics that I've seen so far seem to be fine. It even crossed my mind that Audioguru was referring to the way the transistor is soldered, but there is no photo of it to be found.

 

You might have the collector and emitter of the transistor reversed

<Emphasis mine.>

He doesn't know for sure. He is hypothesizing that the TS might have made a common mistake.

ak

 

These last posts are not anymore about the UV Led circuit, are they? You are talking yet about the normal Leds circuit, yes?
Not sure if this was a good idea to have merged the 2 threads (I created a new thread for the UV Led circuit, but someone merged these 2 threads).

 

Only the voltage drop might be different, between UV and white LEDs. Some modern white LEDs use a blue LED to excite the phosphor.

 

Only the voltage drop might be different, between UV and white LEDs. Some modern white LEDs use a blue LED to excite the phosphor.

I'm using a different circuit and different components. That's why I created a new thread in the first place.

Anyway, I have already build the thing. But I was making a mistake and the UV leds were very dim. I added a push button an a 3.5 jack to the circuit but I was using a resistor with the push button for protection but I was probably placing it in the wrong place (in series between the VCC and the push button, 2.2kΩ), so I think there was a significant voltage drop there and then, not enough for the rest of the circuit, so I decided to remove it.

In the weekend, I'll cut the strip board with the dremel and I can call it complete!

 

Just to call it complete, I leave some screenshots. The LEDs are (trying) to curate 2 drops of solder mask. This solder mask is a bit thick so it might take some time!

 

I recognize some Portuguese words. Are you from Portugal?

Anyway, it seems that you are using blue LEDs to cure solder mask, and I think the wavelength is too short, or perhaps the camera is giving me the wrong impression of reality. You should be using the highest UV possible, which is UV A, as far LEDs are concerned.

 

I recognize some Portuguese words. Are you from Portugal?

Anyway, it seems that you are using blue LEDs to cure solder mask, and I think the wavelength is too short, or perhaps the camera is giving me the wrong impression of reality. You should be using the highest UV possible, which is UV A, as far LEDs are concerned.

I am Portuguese, yes. From Portugal.

And I'm not using blue LEDs, I'm using UV Leds. I think I posted the link of the Leds I purchased somewhere above in this thread!

 

Well, you couldn't be a Portuguese from Spain, unless you were an immigrant.

Regarding the LEDs, I was in doubt, despite the fact that you posted the link, because I don't trust Aliexpress. Note that 365nm should not be visible in any case, and 395nm should be violently violet. Your LEDs appear to be blue.

 

Well, you couldn't be a Portuguese from Spain, unless you were an immigrant.

Regarding the LEDs, I was in doubt, despite the fact that you posted the link, because I don't trust Aliexpress. Note that 365nm should not be visible in any case, and 395nm should be violently violet. Your LEDs appear to be blue.

I mean that I could be a Portuguese/Brazilian. But I'm not. You know the locals pt_pt and pt_br, right? That's why I specified Portuguese from Portugal!

About the LEDs, well, if they lied to me about the LEDs, what can I do, right?
Anyway I can confirm about the wavelength of these LEDs? I mean with tools one may have at home?

 

Yup, forgot about Brazilians. Sorry!

There is a trick you can do if you have a fluorescent marker or, better yet, some solid laundry detergent. Or, even better, a bank note. Basically, if any of those items really fluoresces, your LEDs are emmiting in UV. However, since the camera may be misleading, how do they appear to you? Blue or violet? I ask this because UV LEDs should never be able to emit in blue, unless they are overdriven (by a lot). Also, check the voltage drop, as that is normally corresponds to the wavelength that the LED normally emits (unless overdriven).

 

Yup, forgot about Brazilians. Sorry!

There is a trick you can do if you have a fluorescent marker or, better yet, some solid laundry detergent. Or, even better, a bank note. Basically, if any of those items really fluoresces, your LEDs are emmiting in UV. However, since the camera may be misleading, how do they appear to you? Blue or violet? I ask this because UV LEDs should never be able to emit in blue, unless they are overdriven (by a lot). Also, check the voltage drop, as that is normally corresponds to the wavelength that the LED normally emits (unless overdriven).

Ok, I'll do that when I get home. The emitted light seems violet but I also thought about that. UV shouldn't be visible.
Anyway, I have a total current of about 63mA (from the power supply display, will be a bit more if I use the multimeter) going through 3 parallel branches of 3 LEDs in series, so, about 20mA per branch. Which is the current I got from generic datasheet for 5mm UV LEDs.

 

The current seems to be well within specs. So, that really eliminates that hypothesis. Many LEDs can be driven with 30mA without issues.

 

AliExpress honestly says the cheap LEDs are "UV Purple" and the wavelength numbers go from UV to purple.
Then some of the 100 are UV and others are purple.
Maybe the "UV" ones simply look the same as burnt out LEDs?