Off topic, but it has an orange display.. Heathkit ID-1590 anemometer

[Kerry Borgne]

(See attached)

 I have this unit and the one complaint I have is that it goes through the direction bulbs and they are getting hard to find. I'd like to substitute LEDS for the bulbs (observing polarity or course) and wanted to seek your advice. They would be running on pulsating DC, unfiltered and was concerned about life span. I don't know if LEDS would tolerate this.  Also, from what I can see, I could also eliminate diodes 1 through 8 as the LEDS would also serve that purpose (?) There is never more than 2 leds lit at one time, sometimes simply one.

Your thoughts appreciated.

Kerry

[gregebert]

As long as you dont exceed the RMS current rating of the LED, they will outlast any other type of illuminating device being replaced, especially incandescent.

From the schematic, there are already diodes in the path, so if an LED was installed backwards there's no possible way it can get damaged from reverse-polarity. However, the LED wont illuminate if backwards.

 LEDs have roughly a 2V drop, so you *will* need to add a series resistor; just need to calculate the rough value. The supply is 6.3 VAC (about 9V peak) , and there are 2 diodes in the path. So, the resistor's voltage drop must be about (9-2-0.7 - 0.7 = 5.6 volts). If you get larger diodes that operate at 20mA (should be a good replacement for incandescent lamp), your resistor will be around 270 ohms. Use a 1/4 watt resistor. Obviously you will need to experiment with a few different resistor values to get the desired brightness, so try some values between 220 and 470 ohms.

[jf...@my-deja.com]

Modern LEDs may be "better", but back in the old days, LEDs were no characterized for reverse operation and were pretty poor rectifiers (I forget whether it was leakage or low and unpredictable breakdown voltage, or both).  I would keep the series Si diodes.

 

There was a lot of discussion about multiplexng back in the early days of LED 7-segemnt displays.  In those days, pulsing was preferred since the efficiency went up with the instantaneous drive current so I think I is OK to run the LEDs on unfiltered pulsing dc. 

[Bill van Dijk]

From time to time I read dire warnings about the death of a LED by reverse polarity. It does not happen. A LED is a diode, and is used normally in forward bias, with a resistor (or some other circuit) to limit the current. If the LED is reversed, it simply blocks the current, and precisely nothing happens. You can make a cool bridge rectifier with 4 LEDs, but although technically OK, that is not very practical in most applications due to their high forward drop and low current tolerance. Two wire, bi-color LEDS use this principle where two different color dies are connected back to back, depending on the polarity, one color or the other lights. A red and green LED back to back make therefore a great polarity indicator. Interestingly, if fed an AC current, a third color can be produced, i.e. a green / red combination will shine yellow on AC. RGB LEDs have 3 dies, but can display many colors by lighting one or more dies at varying intensity. This is done through PWM, something a LED is very much suited for. Dimming of LEDs, as well as multiplexing of LEDs is predicated on PWM. Many of the newer Ultra-bright LEDs will still light pleasantly on 5V with series resistors as high as 1.5 kOhm.

The series diode does nothing to protect the LED, but does indeed drop the voltage slightly due to its forward drop. From an efficiency perspective it makes no difference, the diode also dissipates the drop Voltage times the current in heat, just as the resistor does. Since the circuit has a diode it is probably fine to leave it there, but adding one in a new design has no use at all.

 

My apologies to most on the list since you already know all of this. I just felt I had to respond here since I still see this incorrect proposition about reversed LEDs too often.

 

Bill

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.
To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/6ccc7136-e272-442c-a81e-43f93556783c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

This email has been sent from a virus-free computer protected by Avast. www.avast.com

[David Forbes]

Bill,

Thanks for the long rant, but typical LEDs have an absolute maximum
reverse voltage rating of 5V. Exceeding this is not guaranteed to
destroy the LED, but The LED is not guaranteed to work properly after
such use.

It is irresponsible to recommend that people ignore the datasheet
ratings of components.

On 3/24/16 10:35 AM, Bill van Dijk wrote:
>>From time to time I read dire warnings about the death of a LED by reverse polarity. It does not happen. A LED is a diode, and is used normally in forward bias, with a resistor (or some other circuit) to limit the current. If the LED is reversed, it simply blocks the current, and precisely nothing happens. You can make a cool bridge rectifier with 4 LEDs, but although technically OK, that is not very practical in most applications due to their high forward drop and low current tolerance. Two wire, bi-color LEDS use this principle where two different color dies are connected back to back, depending on the polarity, one color or the other lights. A red and green LED back to back make therefore a great polarity indicator. Interestingly, if fed an AC current, a third color can be produced, i.e. a green / red combination will shine yellow on AC. RGB LEDs have 3 dies, but can display many colors by lighting one or more dies at varying intensity. This is done through PWM, somethin
g a LED i
s very much suited for. Dimming of LEDs, as well as multiplexing of LEDs is predicated on PWM. Many of the newer Ultra-bright LEDs will still light pleasantly on 5V with series resistors as high as 1.5 kOhm.
>
> The series diode does nothing to protect the LED, but does indeed drop the voltage slightly due to its forward drop. From an efficiency perspective it makes no difference, the diode also dissipates the drop Voltage times the current in heat, just as the resistor does. Since the circuit has a diode it is probably fine to leave it there, but adding one in a new design has no use at all.
>
>

--
David Forbes, Tucson AZ

[JohnK]

Early LEDs had even lower ratings iirc.

I can remember flicking a design back where two LEDs were in parallel but
opposite ways around for ac operation. The fwd volts of the illuminated one
exceeded the specs for expecting long life of the other one.

John K

----- Original Message -----
From: "David Forbes" <dfo...@dakotacom.net>
To: <neoni...@googlegroups.com>
Sent: Friday, March 25, 2016 4:24 AM
Subject: Re: [neonixie-l] Re: Off topic, but it has an orange display..
Heathkit ID-1590 anemometer

--
You received this message because you are subscribed to the Google Groups
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an
email to neonixie-l+...@googlegroups.com.

To post to this group, send an email to neoni...@googlegroups.com.

To view this discussion on the web, visit

https://groups.google.com/d/msgid/neonixie-l/56F429CC.2040003%40dakotacom.net.

[gregebert]

In the spirit of nixies, though, you could redo the entire design and use NE-2 bulbs. I've driven NE-2's with opto-triacs on a few projects, which allows you to safely (ie, with proper isolation) control AC-line voltages from digital logic. The MOC3021 has 5kV isolation and costs about 50 cents.

I cant see the rest of the schematic, so I would assume it's NOT safe to replace 6.3VAC with 120VAC, and replace the bulbs with NE-2 bulbs that have a built-in resistor. Besides, the bulbs would be running on DC, rather than AC so they will appear less-bright with only one electrode glowing, rather than 2.

[Jeff Walton]

I recently purchased some Orange LED’s – 3mm from a vendor on eBay and they are the absolute closest color that I have ever seen to a nixie tube glow.  I used them to replace some yellow colons in a clock kit and they are spot-on with IN-12 display tubes.  I would have preferred neon bulbs but these were drop-in with a resistor change to reduce the brightness.  (they are very bright).  From a distance, they appear the same color as the nixie and are a single light source (not a mix of red and green..)

 

http://www.ebay.com/itm/151952045037?_trksid=p2057872.m2749.l2649&ssPageName=STRK%3AMEBIDX%3AIT

 

I tried several sources before I found one that was close enough in wavelength.  This was really good and they are inexpensive.

 

 

 

From: neoni...@googlegroups.com [mailto:neoni...@googlegroups.com] On Behalf Of Kerry Borgne

Sent: Thursday, March 24, 2016 6:37 AM
To: neonixie-l

--

You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To post to this group, send email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/926dc5ec-17e7-46b8-b478-88f2218c0a5d%40googlegroups.com.

[Bill van Dijk]

David,

Indeed, everything has a limit, and I did not touch on that. An obvious (and significant) oversight. Indeed, many LEDs have a reverse voltage limit of around 5V, and as such my statements with respect to 5V logic circuits holds true. But that is not the end of the story either.

Some high-end LEDs actually have 2 back to back zener diodes (parallel with the LED die) to protect them against reverse power. For instance, the Everlight 334-15/T1C1-4WYA white LED also has a 5V reverse voltage limit, but allows a 100mA reverse zener current. That means that if the series resistor limits the forward current to its forward limit of 20 to 30mA, the reverse installation would be able to withstand whatever voltage is used in the "normal" forward position.

Most LEDs do not have that kind of protection (although the designer could build in an external protection zener) and will not tolerate much more than 10uA or so in reverse.

LEDs were not designed for reverse operation outside the 5V logic levels, and as such vary a lot as you already pointed out. For example, the Vishay TLUR6400 has a typical (guaranteed) reverse max of 6V, with a max value of 15V.

Thanks for pointing that out.

Bill

--
You received this message because you are subscribed to the Google Groups "neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email to neonixie-l+...@googlegroups.com.

To post to this group, send an email to neoni...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/neonixie-l/56F429CC.2040003%40dakotacom.net.

For more options, visit https://groups.google.com/d/optout.

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

[Kerry Borgne]

So, installed the LEDS and all is well.... almost. One of the reed switches in the outdoor sending unit (you know, the one up on the roof!)  is bad, plus there appears to be a bit of leakage on the senders internal circuit board as on occasion an LED will light very dimly when it shouldn't. Not too surprising as it's been up there for at least 15 years!  I guess a trip up onto the roof is in order AFTER the weather settles down..

[GastonP]

The LEDs could light very dimly due to, as you say, leakage on the "sender", but it also could do so due to mechanical vibration of the system in the roof, spinning of the wind arrow or to leakage in the cable.
If it is not too disturbing, just don't run the risk. Remember too that the circuit was designed to run lamps which are less sensitive to pulses and leakage.

[GastonP]

Also, having the schematic, you can test for leakages without need of climbing to the roof :)