LED wiring questions (parallel/series)

Started by dave999z, February 27, 2018, 11:13:39 PM

Previous topic - Next topic

dave999z

Hello -

I'm building a true bypass looper with 14 switches.  Therefore, it will have 14 LEDs.  They will be wired in parallel, obviously, so that turning one off doesn't turn all the others off.  Each LED will have its own series resistor.

My first question is...  Assuming you're using 14 identical LEDs and series resistors, will the brightness of all the LEDs vary somewhat depending on how many are active.  For example, if only 1 effect/LED is turned on, is that single LED going to glow brighter than each LED will glow when I have all 14 effects/LEDs active?  In other words, the more parallel LEDs are active, the dimmer each LED becomes?  If the answer is yes, is there any way around that?

My second question is...  If I add one additional "master" resistor upfront, before any of the parallel branches, would that create/exacerbate the above problem?  (The reason I want to have a master resistor upfront is so that if I want to change the brightness of all the LEDs I would just need to swap out that master resistor, as opposed to swapping out all 14 resistors.)

Thanks in advance for any help
-dave

Rob Strand

#1
Quote
My first question is
Yes, in general it's not really practical to use a series resistor if the number LEDs is varying.

This problem is solved using a constant current source instead of a resistor.  It was a common trick
used for bar graph displays.

How to switch series individual LED in a series connection has its own complications.  The bargraphs divide between a set of on LEDS and a set of OFF leds using a fairly simple switching scheme.

QuoteMy second question is... 
In general any semiconductor connected in parallel will have some form of current sharing problem.  In practice the parts are not the exactly same and unless specially matched parts are used, like those inside of IC's, the degree of matching is quite poor.  *Sometimes* when you can parallel devices that have a lot of internal resistance.  The internal resistance takes up the slack.  LEDs are on the boarder.  You might get away with it but then again you might find some are dim and some don't light at all.

Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.

dave999z

Quote from: Rob Strand on February 27, 2018, 11:33:46 PM
Yes, in general it's not really practical to use a series resistor if the number LEDs is varying.

I’m sorry I don’t really follow that.

So the answer is yes the brightness of each LED will vary based on how many other parallel resistor/LED pairs are active?

When you say not practical to use a series resistor... are you referring to the series resistor within each parallel branch?  I haven’t seen people wire LEDs in true bypass loopers any other way, which leaves me confused.

PRR

> Assuming you're using 14 identical LEDs and series resistors, will the brightness of all the LEDs vary somewhat depending on how many are active.

I have 14 lights in my house. Does the first one get dim if I turn-on 13 more?
(In my house: yes, slightly.)

If the supply voltage is SOLID, 9.0V from no-load to HUGE load, then no variation. Actually if it is steady 20% from zero LEDs to 14 LEDs, your eye will not notice.
(I live too far from the street and my "120V" sags as load is added.)

You "most certainly can" wire in series, assuming un-grounded switching. Worked on arc-lights for decades. Still does for some runway lights. You need a steady current (instead of steady voltage). Short each LED, or not, to darken or glow it. This is probably not practical for fourteen 2V-3V LEDs and our customary 9V pedal supply. If a system has a raw 36V DC supply in it, this becomes very economical. (10-LED VU displays in 24V systems are generally run series.)
  • SUPPORTER

Rob Strand

QuoteI'm sorry I don't really follow that.

So the answer is yes the brightness of each LED will vary based on how many other parallel resistor/LED pairs are active?

When you say not practical to use a series resistor... are you referring to the series resistor within each parallel branch?  I haven't seen people wire LEDs in true bypass loopers any other way, which leaves me confused.
Sorry, lets back pedal a bit.   I think I misunderstood series setup as having one series resistor for the entire thing.
PRR understood you correctly.

If each LED has a it's own resistor then you don't have a problem.   The brightness of each LED will the virtually independent of the others.    Virtual only in the sense of the battery drooping which is a very small effect.

QuoteMy second question is...  If
Yes, it makes it worse.  Here the you are creating a situation to promote the droop problem.

Is there a reason you want to do that?
Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.

dave999z

Quote from: Rob Strand on February 28, 2018, 01:00:29 AM
Is there a reason you want to do that?

Yes, from my first post up above...

The reason I want to have a master resistor upfront is so that if I want to change the brightness of all the LEDs I would just need to swap out that master resistor, as opposed to swapping out all 14 resistors.

Ice-9

Your second question..
If I wanted to have a master brightness to control the overall level on ALL LED's then instead of a master resistor in series with the 14 other LED resistors I would use an adjustable voltage regulator which must be capable of providing enough current to all 14 resistor/LED networks. This will allow all 14 LED's to keep the same brightness if 1 or 14 are turned on and also allow the overall brightness to adjusted on ALL LED's.
Note Your Power supply must also be able to provide enough current for all this AND anything else that is powered from the power supply.
www.stanleyfx.co.uk

Sanity: doing the same thing over and over again and expecting the same result. Mick Taylor

Please at least have 1 forum post before sending me a PM demanding something.

antonis

As Mick above said, you need a variable voltage source, provided it can supply constant current..  :icon_wink:

You may consider each LED with its own series current limiting resistor as a "unity" which glows in respect with current flowing through it..
For a lower voltage across it, less current will flow hence less LED brightness...
(provided constant LED forward voltage drop, current flowing is proportional to (VSupply - VLed drop) / Rseries..)
"I'm getting older while being taught all the time" Solon the Athenian..
"I don't mind  being taught all the time but I do mind a lot getting old" Antonis the Thessalonian..

dave999z

Thanks everyone.  I don't need an actual brightness control.  Was just wondering if sticking one extra master resistor upfront would cause the brightness to vary based on how many LEDs are currently active, and it sounds like the answer is yes.  So I'll just skip doing that.  I'll pick a value for all the parallel resistors now that gives me a brightness level I like and live with it.

Ice-9

Quote from: dave999z on February 28, 2018, 11:43:16 AM
Thanks everyone.  I don't need an actual brightness control.  Was just wondering if sticking one extra master resistor upfront would cause the brightness to vary based on how many LEDs are currently active, and it sounds like the answer is yes.  So I'll just skip doing that.  I'll pick a value for all the parallel resistors now that gives me a brightness level I like and live with it.

In that case just wire up a test LED and select the resistor value that give the brightness you want then use the 14 resistors of that value for all 14 LED, jobs a good-un.
www.stanleyfx.co.uk

Sanity: doing the same thing over and over again and expecting the same result. Mick Taylor

Please at least have 1 forum post before sending me a PM demanding something.