how to determine clr value

[bobboxbody]

I'm new to diy pedals and just built a dead end fx Flint, but I used a 3mm blue led instead of a 5mm red led because I had one left over from my first pedal build. The current limiting resistor for the red led is 1.5k, how do I figure out the correct value for blue? I failed at googling and answer, please don't hate me and thanks in advance for your help.

[PRR]

Welcome.

Try the specified value. The current will be a bit low but that's no big deal. See if it needs to be brighter. Because the general improvement in LEDs, it may be too bright; use a bigger resistor, 2k, 3k.

[bobboxbody]

Welcome.

Try the specified value. The current will be a bit low but that's no big deal. See if it needs to be brighter. Because the general improvement in LEDs, it may be too bright; use a bigger resistor, 2k, 3k.

Thanks, will do.

[bluebunny]

Do what Paul says.  BTW, there are online calculators for LED current-limiting resistors, but note that these will tell you the minimum value beyond which the LED will fry.  This minimum is likely to be very bright indeed - painfully so.  (Ask me how I know.  Heathrow weren't happy with me that evening...)  The real-world value is going to be much higher than the calculator tells you.

[GibsonGM]

One thing you can do is determine the min. resistance to keep the LED safe, and add a pot (wired as a variable resistor) to that...this will let you dial in LED brightness just where you want it.

When the LED looks like you want it to, you remove the pot, measure resistance across the end and wiper lugs, and add that to the protection resistor value.  Try a 10k pot; modern resistors often need high-ish R values, they are so efficient. 

[ElectricDruid]

Just for completeness, here's what the calculators are doing. It's simple Ohm's Law.

You need to know two things about the LED: The forward voltage, V_f, and the maximum current, I_f. You also know what supply voltage you're using to light the LED.



Now, since the only two parts in this circuit are the LED and the resistor, we know that if we've dropped Vf across the LED, the rest of the voltage must be dropped across the resistor.
E.g. The voltage across the resistor V_r is the supply voltage V_s minus the LED forward voltage:
V_r = V_s - V_f
We also know that if a currrent of I_f is flowing through the LED, it must be flowing through the resistor too. That gives us two out of three facts, voltage and current, so then we can use Ohm's Law V/I=R to find the resistance:
V_r / I_f = ( V_s - V_f ) / I_f = R

If we do a quick example with an LED that has a V_f of 2V and a maximum current I_f of 20mA running on a 9V supply:
(9 - 2) / 0.020 = 7 / 0.020 = 350R
So any resistor from 350R or bigger will be safe for that LED. In practice, we might well decide we only want 5mA LED current, to save juice and make LED less retina-burning. We can work out the resistor for that too, the same way:
(9 - 2) / 0.005 = 7 / 0.005 = 1400 = 1K4
So 1K5 would be the value we'd use, since 1K4 resistors aren't common.

[GibsonGM]

Only issue I've ever had with that is not having the Vf. Back in the RS days, they'd give you that on the package. Now we often have to measure it if we want it...my meter can't, it doesn't have the range. So I still guess at the CLR, rather than 'experiment'.  Since the 'new' ultra brights came out, 2.2k is a fine safe place to start and can use the pot trick to find correct brightness.

I'm old enough to remember DOING that calculation before blowing out a precious LED, ha ha!  I still DID once in a while, tho   

[bobboxbody]

Thanks for all the replies!
The calculators and ohm's law were what I found in my googling, but I assumed I didn't want the minimum value and I was too ignorant/bad at math to translate the current value with red forward voltage to the correlating value with blue voltage using ohm's law.
I finished the build with the default value and the led is indeed too bright, also I can't get the correct voltages in Q2 even after trying a bunch of combinations of 4 different 1N4148, BAT46, and LND150N3. I socketed the spots for Q1, Q2, D1, and D2 and then kept swapping in the extra diodes and transistors, the second swap got Q1 close enough, but Q2 is still about 1V low.
Could the CLR be effecting(affecting?) the voltage of the LND150N3? It doesn't look like it should in the schematic, but I'm bad at understanding schematics.

Edit: Just bumped clr up to 3.9k and it's reached an acceptable level, only semi-blinding. No change in Q2 voltages, so I guess it's not related.

[ElectricDruid]

Here's the build doc with schematic if anyone wants to answer that question:

https://drive.google.com/file/d/15-5mA6tFOJ0HPRaLysr-KKzkoP7cWn3w/view

Bobboxbody, are you talking about D4/R10 on the schematic? Or is there an offboard CLR+LED hanging off the 3PDT switch?

[bobboxbody]

Here's the build doc with schematic if anyone wants to answer that question:

https://drive.google.com/file/d/15-5mA6tFOJ0HPRaLysr-KKzkoP7cWn3w/view

Bobboxbody, are you talking about D4/R10 on the schematic? Or is there an offboard CLR+LED hanging off the 3PDT switch?

Yes, D4/R10, wasn't sure if I was allowed to post the schematic, thank you.

[digi2t]

Always a dilemma when I trace them out. Do I post the original value, or just "CLR", since I have no clue what a builder will use for the LED.

Typically, I just use whatever value so the LED doesn't blind me when I turn it on. I find 5mm clear blue LED's are the worst, so I usually use a 10k with those, otherwise I'm seeing blue spots for hours afterwards.

[bobboxbody]

Always a dilemma when I trace them out. Do I post the original value, or just "CLR", since I have no clue what a builder will use for the LED.

Typically, I just use whatever value so the LED doesn't blind me when I turn it on. I find 5mm clear blue LED's are the worst, so I usually use a 10k with those, otherwise I'm seeing blue spots for hours afterwards.

I'll bump it up to 10k, still too bright for me. Any tips on Q2 voltage‽ Is it dependent on both diodes or just the bat46? I have a fairly close combo now for Q1, just far off on Q2, but it sounds good anyway.

[antonis]

A: Q2 Drain voltage higher than 3.7V
Try either raising Q2 V_GS (by using a diode of higher forward voltage drop instead of BAT46) or raise R6+R7 total resistance value..

B: Q2 Drain voltage lower than 3.7V
Retain BAT46 and lower R6+R7 total resistance value..

[Ben N]

Or just rig yourself up a little tester and measure: wire a 25k-50k pot up as a variable resistor, with a small resistor (470R-1k) in series as your min value, a battery clip or power jack, and a couple of alligator clip leads. Clip up your LED (in the right direction, natch), and twist the pot around til you're happy with it. Measure the resistance across the pot/resistor combo, find your closest matching value, and there you are: a customized R-LED. If you use a power jack instead of a battery, you might want to include your preferred polarity protection/power filtering as well to get the voltage just right, but it probably doesn't matter much.

[amptramp]

If you have access to a resistive photoconductor (and you might not because they are usually cadmium sulphide or cadmium selenide, which violates RoHS rules), you can put it in series with the limiting resistor to reduce the current at low ambient light levels.  You can add a parallel resistor to ensure you have enough luminance when the ambient light is low.  You can find photoresistors with a high-brightness resistance of several hundred ohms and dark resistance of several hundred Kohms.  There is not going to be a single fixed resistor that fits all applications.  You could be playing an outdoor concert one day and a completely dark venue the next day.

If you can't get photoresistors or you intend to sell your product in Europe where RoHS is enforced, you can get sun battery photovoltaic cells or photodiodes you can use to bias a transistor that controls LED current.

[Ben N]

It would actually not be entirely nuts to include a trimpot in series with a min value resistor in the final product, or together with Ron's photoresistor idea. Let the user pick their preferred brightness level.

[Mark Hammer]

I have a little gizmo I made with a 12-position rotary switch, selecting fixed resistors, ranging from more than I likely need (given the efficiency of the brightest LEDs in the parts drawer), to various lower values.  I start from the highest value, and turn the switch downwards, until I get to a desirable brightness.  Since 12 values is much less than the number between the lowest and highest value used, I use the "best" illumination as a guide to selecting common resistor value a little higher or lower.

[ElectricDruid]

It would actually not be entirely nuts to include a trimpot in series with a min value resistor in the final product, or together with Ron's photoresistor idea. Let the user pick their preferred brightness level.

Watch out for the power rating of the trimpot if you try this. Many are only 0.1W, and that's 20mA at 5V. So with Vf=2V and a 9V supply, you'd fry it. There are plenty of 0.2W and 0.5W trimmers too though, so it's not a limitation, just a gotcha.

[Mark Hammer]

I'll just add that brightness is not identical to visibility.  Brightness assists visual contrast, which dictates visibility.  One needs to consider the background against which the LED will be shining.  So, for instance, a red or blue status LED, in a black enclosure, wouldn't have to be as bright as a yellow LED against a beige or white enclosure.  Black plastic bezels can serve to make LEDs nicely visible, regardless of enclosure surface/colour, and shiny chrome ones can require a little more illumination for better contrast.

In past, I've used a single LED as both an indicator of rate and effect status.  The stomp switch selects the current-limiting resistor value going to the LED.  In effect mode, the resistance is lower than bypass mode, such that relative brightness tells me the effect status, but I can still use blink rate to set modulation speed, even in bypass.  As you can imagine, selecting the "right" CLR value for the two modes can be tricky.

[bobboxbody]

A: Q2 Drain voltage higher than 3.7V
Try either raising Q2 V_GS (by using a diode of higher forward voltage drop instead of BAT46) or raise R6+R7 total resistance value..

B: Q2 Drain voltage lower than 3.7V
Retain BAT46 and lower R6+R7 total resistance value..

r6 seems to be only changing Q1 value with r7 and r8 jumpered, r9 needs adjustment for Q2, I got there eventually after banging my head against the wall. Thanks for all the replies everyone.