Indicator Lamp on Weathertron 3AAT80B1A1 Thermostat

Posted by Fartikus on November 17, 2008, 11:13 pm

>>> Jeff Wisnia wrote:
>>>> kool wrote:
>>>>>> One of the GE Weathertron Model 3AAT80B1A1 thermostats in our home is
>>>>>> doing its thing and when the temperature setting is increased a few
>>>>>> degrees above ambient the auxillary heaters come on OK to supplement
>>>>>> the heat pump's output.
>>>>>> The blue "Aux Heat" indicator recently stopped lighting in that mode.
>>>>>> I could live with it that way, but being an engineer I'm sort of anal
>>>>>> about having everything working the way it was designed to.
>>>>>> I suspect it's just a "burned out bulb". The thermostat is old enough
>>>>>> so I'd doubt that the indicators use LEDs.
>>>>>> Before I approach it, can someone tell me if replacement lamps are
>>>>>> available and easily screwed or snapped in place or will I have to
>>>>>> figure out what tiny bulb to buy and use my past years of electronic
>>>>>> repair experience to solder it in.
>>>>>> It would be annoying to toss out an otherwise working thermostat just
>>>>>> for the want of a bulb.
>>>>>> Thanks guys,
>>>>>> Jeff
>>>>>> --
>>>>>> Jeffry Wisnia
>>>>>> (W1BSV + Brass Rat '57 EE)
>>>>>> The speed of light is 1.8*10^12 furlongs per fortnight.
>>>>> It's a tiny incandescent bulb about the size a Motrin gel-cap , with
>>>>> two little wires protruding from the bottom G.E.# 35E. It just sits in
>>>>> the hole over a blue plastic lens.(the bulb is clear.) Mounts only by
>>>>> the wires soldered onto the wires in the stat.
>>>> I soldered in a new bulb yesterday. Tracing the circuit showed that
>>>> here's a single diode in the common return lead of both bulbs so the
>>>> bulbs effectively see only half of the nominal 24 volt AC supply.
>>>> I used a 12-14 volt "grain of wheat" bulb and it works fine now.
>>>> Jeff
>>> Now that would be RMS voltage and that diode would indicate the bulb is
>>> DC rather then AC. basically that bulb has a half cycle to cool and a
>>> half cycle the heat and produce light. If all of that is not needed,
>>> redesign and install a lifetime LED.

>> Here is an explanation how it works.
>> The bulb would get approx 12 VDC pulses and the 0.7v junction voltage of
>> a (silicon) diode would reduce the actual by 0.7V
>> http://www.wisc-online.com/objects/index_tj.asp?objID=SSE402
>> By the way, a grain of wheat bulb is incandescant, so it does not matter
>> whether you feed it AC or DC. If an LED wre used (with a current limiting
>> resistor to reduce the voltage to the LED to its rating, feeding it the
>> pulsating approx 12VDC would be fine as long as it was wired for the
>> correct polarity.

> On this application there is no DC involved. The power to the two bulbs
> comes directly off of the "B" terminal on the stat through the diode to
> one wire from each bulb. the other wire from the blue bulb (auxillary
> heat) goes to the "U" terminal and the second wire from the red bulb
> (emergency heat) goes to the "F" terminal. The F terminal is not fan as it
> also has a G terminal as well as Y,X2,W, O,R and T. on the BAY subbase.
> 1981 vintage no electronics other than 1 diode.

Kool, the original poster Jeff Wisnia, as quoted above is the one that said
he traced a single diode in the return leg of the 2 bulbs. The diode is
built into the thermostat and half wave rectifies the 24VAC used in the HVAC
control supply.

Just for fun today I looked up what a grain of wheat bulb is rated at. I
found at least one that is rated at 75 milliamps at 12 volts.

I hooked up a transformer that has a 28 volt secondary, measured it to
actually be 30.5 VAC and hooked up a 1N4007 ( a common switching diode) in
series with 3 120 ohm resistors (total 360 ohms). That 360 ohms across the
30.5 volts AC in series with the diode draws .084 amps, similar to the .075
amp rating of the grain of wheat bulb.

I measured the voltage across the resistors total resistance and read 13.3
volts DC with a brand new Fluke meter. 13.3 volts will not significantly
reduce the life of a 12 volt bulb. I did not measure the voltage with a
scope. All light bulbs are rated at a voltage. This voltage rating is the
RMS voltage. All AC voltage does have a peak to peak value which works out
to be 2.8 the RMS voltage. , yet the 120 volt bulb lasts for years.

Mark said the voltage to the bulb would be 16.5 volts, but that is the 0
point to 12 volt RMS level x 1.4 peak multiplier math. The half wave
rectified voltage to the bulb with the bulb loading this down somewhat
measures out at 12 volts DC. If I get a chance tomorrow, I will hook up a
scope and see what the peak voltage levels to the load is, just out of
curiosity.

Posted by Jeff Wisnia on November 18, 2008, 12:22 am

Fartikus wrote:

>>>>Jeff Wisnia wrote:
>>>>>kool wrote:
>>>>>>>One of the GE Weathertron Model 3AAT80B1A1 thermostats in our home is
>>>>>>>doing its thing and when the temperature setting is increased a few
>>>>>>>degrees above ambient the auxillary heaters come on OK to supplement
>>>>>>>the heat pump's output.
>>>>>>>The blue "Aux Heat" indicator recently stopped lighting in that mode.
>>>>>>>I could live with it that way, but being an engineer I'm sort of anal
>>>>>>>about having everything working the way it was designed to.
>>>>>>>I suspect it's just a "burned out bulb". The thermostat is old enough
>>>>>>>so I'd doubt that the indicators use LEDs.
>>>>>>>Before I approach it, can someone tell me if replacement lamps are
>>>>>>>available and easily screwed or snapped in place or will I have to
>>>>>>>figure out what tiny bulb to buy and use my past years of electronic
>>>>>>>repair experience to solder it in.
>>>>>>>It would be annoying to toss out an otherwise working thermostat just
>>>>>>>for the want of a bulb.
>>>>>>>Thanks guys,
>>>>>>>Jeff
>>>>>>>--
>>>>>>>Jeffry Wisnia
>>>>>>>(W1BSV + Brass Rat '57 EE)
>>>>>>>The speed of light is 1.8*10^12 furlongs per fortnight.
>>>>>>It's a tiny incandescent bulb about the size a Motrin gel-cap , with
>>>>>>two little wires protruding from the bottom G.E.# 35E. It just sits in
>>>>>>the hole over a blue plastic lens.(the bulb is clear.) Mounts only by
>>>>>>the wires soldered onto the wires in the stat.
>>>>>I soldered in a new bulb yesterday. Tracing the circuit showed that
>>>>>here's a single diode in the common return lead of both bulbs so the
>>>>>bulbs effectively see only half of the nominal 24 volt AC supply.
>>>>>I used a 12-14 volt "grain of wheat" bulb and it works fine now.
>>>>>Jeff
>>>>Now that would be RMS voltage and that diode would indicate the bulb is
>>>>DC rather then AC. basically that bulb has a half cycle to cool and a
>>>>half cycle the heat and produce light. If all of that is not needed,
>>>>redesign and install a lifetime LED.
>>>Here is an explanation how it works.
>>>The bulb would get approx 12 VDC pulses and the 0.7v junction voltage of
>>>a (silicon) diode would reduce the actual by 0.7V
>>>http://www.wisc-online.com/objects/index_tj.asp?objID=SSE402
>>>By the way, a grain of wheat bulb is incandescant, so it does not matter
>>>whether you feed it AC or DC. If an LED wre used (with a current limiting
>>>resistor to reduce the voltage to the LED to its rating, feeding it the
>>>pulsating approx 12VDC would be fine as long as it was wired for the
>>>correct polarity.

>>On this application there is no DC involved. The power to the two bulbs
>>comes directly off of the "B" terminal on the stat through the diode to
>>one wire from each bulb. the other wire from the blue bulb (auxillary
>>heat) goes to the "U" terminal and the second wire from the red bulb
>>(emergency heat) goes to the "F" terminal. The F terminal is not fan as it
>>also has a G terminal as well as Y,X2,W, O,R and T. on the BAY subbase.
>>1981 vintage no electronics other than 1 diode.

>
>
> Kool, the original poster Jeff Wisnia, as quoted above is the one that said
> he traced a single diode in the return leg of the 2 bulbs. The diode is
> built into the thermostat and half wave rectifies the 24VAC used in the HVAC
> control supply.
>
> Just for fun today I looked up what a grain of wheat bulb is rated at. I
> found at least one that is rated at 75 milliamps at 12 volts.
>
> I hooked up a transformer that has a 28 volt secondary, measured it to
> actually be 30.5 VAC and hooked up a 1N4007 ( a common switching diode) in
> series with 3 120 ohm resistors (total 360 ohms). That 360 ohms across the
> 30.5 volts AC in series with the diode draws .084 amps, similar to the .075
> amp rating of the grain of wheat bulb.
>
> I measured the voltage across the resistors total resistance and read 13.3
> volts DC with a brand new Fluke meter. 13.3 volts will not significantly
> reduce the life of a 12 volt bulb. I did not measure the voltage with a
> scope. All light bulbs are rated at a voltage. This voltage rating is the
> RMS voltage. All AC voltage does have a peak to peak value which works out
> to be 2.8 the RMS voltage. , yet the 120 volt bulb lasts for years.
>
> Mark said the voltage to the bulb would be 16.5 volts, but that is the 0
> point to 12 volt RMS level x 1.4 peak multiplier math. The half wave
> rectified voltage to the bulb with the bulb loading this down somewhat
> measures out at 12 volts DC. If I get a chance tomorrow, I will hook up a
> scope and see what the peak voltage levels to the load is, just out of
> curiosity.
>
>

Just for Ss&Gs I asked light bulb guru Don Klipstein about the 120 volt
bulb with a diode in series with it connected to a 240 volt AC supply.
His reply was:

*****************************

240V AC halved by an "ideal diode" (zero voltage drop) will have the
same average voltage as unmodified 120V AC. However, the RMS (root
mean square) voltage will be different - higher by a factor of square
root of 2. That would be 169.7 volts. The RMS voltage is the effective
voltage for simple resistive loads such as incandescent lightbulbs and
resistive heaters.

Think of it this way: An ideal resistor (resistance not varying with
temperature) receives 1/4 as much watts when voltage is halved. Not
only is voltage halved, but amps are also halved, and that means watts
are quartered.

But a diode merely disconnects the load from the line for half the time,
so wattage is merely halved. By Ohm's Law, the effective voltage of AC
after a diode is then the original voltage times square root of 1/2.

Incandescent lamps do have the complication of their resistance varying
with temperature, but the effective voltage ("RMS voltage") is still the
same as if resistance was constant. An incandescent lightbulb receiving
240V AC through a diode is effectively getting 169 volts. After the
effect of resistance varying with temperature, power consumption by the
lightbulb will be close to 1.65-1.72 times that at 120V rather than
double.

Keep in mind that a voltmeter will read a voltage other than about
169 volts unless it is a "True RMS" type. It may read 120 if it has
fullwave rectification or possibly either 240 or close to zero due to
having mere halfwave rectification. On a DC range, a non-true-RMS
voltmeter will read halfwave rectified 240V or fullwave rectified 120V
as about 108 volts - the actual average voltage. (Nominal voltage of AC
is the RMS voltage, about 11% higher than the average for a sine wave.)

According to what I consider a "usual 1-size-fits-all" rule, life
expectancy is reduced by a factor of about 60 - and actual life
expectancy results can vary significantly, even greatly.

I would caution that burnout may be more spectacular than at 120V,
possibly even unsafe, especially for a diode unless it is big enough to
not protect your fuse or circuit breaker by blowing first (possibly
explosively) should a "burnout arc" form.

And at 70% overwattage, the bulb may overheat and break - especially in
an enclosed fixture or a recessed ceiling fixture.

Lower wattage 120V lightbulbs, such as maybe around or under 25 watts
(?), may have their filament vary enough in temperature over each AC
cycle when used with a diode and 240V to change results somewhat from
results with actual 169-170V - probably for the worse in terms of life
expectancy.

- Don Klipstein (don@misty.com)

***********************************

That seems to agree with Mark's explanation....

Jeff
--
Jeffry Wisnia
(W1BSV + Brass Rat '57 EE)
The speed of light is 1.98*10^14 fathoms per fortnight.

Posted by Fartikus on November 18, 2008, 6:13 am

> Fartikus wrote:

>>>>>Jeff Wisnia wrote:
>>>>>>kool wrote:
>>>>>>>>One of the GE Weathertron Model 3AAT80B1A1 thermostats in our home
>>>>>>>>is doing its thing and when the temperature setting is increased a
>>>>>>>>few degrees above ambient the auxillary heaters come on OK to
>>>>>>>>supplement the heat pump's output.
>>>>>>>>The blue "Aux Heat" indicator recently stopped lighting in that
>>>>>>>>mode. I could live with it that way, but being an engineer I'm sort
>>>>>>>>of anal about having everything working the way it was designed to.
>>>>>>>>I suspect it's just a "burned out bulb". The thermostat is old
>>>>>>>>enough so I'd doubt that the indicators use LEDs.
>>>>>>>>Before I approach it, can someone tell me if replacement lamps are
>>>>>>>>available and easily screwed or snapped in place or will I have to
>>>>>>>>figure out what tiny bulb to buy and use my past years of electronic
>>>>>>>>repair experience to solder it in.
>>>>>>>>It would be annoying to toss out an otherwise working thermostat
>>>>>>>>just for the want of a bulb.
>>>>>>>>Thanks guys,
>>>>>>>>Jeff
>>>>>>>>--
>>>>>>>>Jeffry Wisnia
>>>>>>>>(W1BSV + Brass Rat '57 EE)
>>>>>>>>The speed of light is 1.8*10^12 furlongs per fortnight.
>>>>>>>It's a tiny incandescent bulb about the size a Motrin gel-cap , with
>>>>>>>two little wires protruding from the bottom G.E.# 35E. It just sits
>>>>>>>in the hole over a blue plastic lens.(the bulb is clear.) Mounts only
>>>>>>>by the wires soldered onto the wires in the stat.
>>>>>>I soldered in a new bulb yesterday. Tracing the circuit showed that
>>>>>>here's a single diode in the common return lead of both bulbs so the
>>>>>>bulbs effectively see only half of the nominal 24 volt AC supply.
>>>>>>I used a 12-14 volt "grain of wheat" bulb and it works fine now.
>>>>>>Jeff
>>>>>Now that would be RMS voltage and that diode would indicate the bulb is
>>>>>DC rather then AC. basically that bulb has a half cycle to cool and a
>>>>>half cycle the heat and produce light. If all of that is not needed,
>>>>>redesign and install a lifetime LED.
>>>>Here is an explanation how it works.
>>>>The bulb would get approx 12 VDC pulses and the 0.7v junction voltage of
>>>>a (silicon) diode would reduce the actual by 0.7V
>>>>http://www.wisc-online.com/objects/index_tj.asp?objID=SSE402
>>>>By the way, a grain of wheat bulb is incandescant, so it does not matter
>>>>whether you feed it AC or DC. If an LED wre used (with a current
>>>>limiting resistor to reduce the voltage to the LED to its rating,
>>>>feeding it the pulsating approx 12VDC would be fine as long as it was
>>>>wired for the correct polarity.
>>>On this application there is no DC involved. The power to the two bulbs
>>>comes directly off of the "B" terminal on the stat through the diode to
>>>one wire from each bulb. the other wire from the blue bulb (auxillary
>>>heat) goes to the "U" terminal and the second wire from the red bulb
>>>(emergency heat) goes to the "F" terminal. The F terminal is not fan as
>>>it also has a G terminal as well as Y,X2,W, O,R and T. on the BAY
>>>subbase. 1981 vintage no electronics other than 1 diode.

>> Kool, the original poster Jeff Wisnia, as quoted above is the one that
>> said he traced a single diode in the return leg of the 2 bulbs. The diode
>> is built into the thermostat and half wave rectifies the 24VAC used in
>> the HVAC control supply.
>> Just for fun today I looked up what a grain of wheat bulb is rated at. I
>> found at least one that is rated at 75 milliamps at 12 volts.
>> I hooked up a transformer that has a 28 volt secondary, measured it to
>> actually be 30.5 VAC and hooked up a 1N4007 ( a common switching diode)
>> in series with 3 120 ohm resistors (total 360 ohms). That 360 ohms across
>> the 30.5 volts AC in series with the diode draws .084 amps, similar to
>> the .075 amp rating of the grain of wheat bulb.
>> I measured the voltage across the resistors total resistance and read
>> 13.3 volts DC with a brand new Fluke meter. 13.3 volts will not
>> significantly reduce the life of a 12 volt bulb. I did not measure the
>> voltage with a scope. All light bulbs are rated at a voltage. This
>> voltage rating is the RMS voltage. All AC voltage does have a peak to
>> peak value which works out to be 2.8 the RMS voltage. , yet the 120 volt
>> bulb lasts for years.
>> Mark said the voltage to the bulb would be 16.5 volts, but that is the 0
>> point to 12 volt RMS level x 1.4 peak multiplier math. The half wave
>> rectified voltage to the bulb with the bulb loading this down somewhat
>> measures out at 12 volts DC. If I get a chance tomorrow, I will hook up a
>> scope and see what the peak voltage levels to the load is, just out of
>> curiosity.

> Just for Ss&Gs I asked light bulb guru Don Klipstein about the 120 volt
> bulb with a diode in series with it connected to a 240 volt AC supply. His
> reply was:
> *****************************
> 240V AC halved by an "ideal diode" (zero voltage drop) will have the
> same average voltage as unmodified 120V AC. However, the RMS (root
> mean square) voltage will be different - higher by a factor of square root
> of 2. That would be 169.7 volts. The RMS voltage is the effective
> voltage for simple resistive loads such as incandescent lightbulbs and
> resistive heaters.
> Think of it this way: An ideal resistor (resistance not varying with
> temperature) receives 1/4 as much watts when voltage is halved. Not only
> is voltage halved, but amps are also halved, and that means watts are
> quartered.
> But a diode merely disconnects the load from the line for half the time,
> so wattage is merely halved. By Ohm's Law, the effective voltage of AC
> after a diode is then the original voltage times square root of 1/2.
> Incandescent lamps do have the complication of their resistance varying
> with temperature, but the effective voltage ("RMS voltage") is still the
> same as if resistance was constant. An incandescent lightbulb receiving
> 240V AC through a diode is effectively getting 169 volts. After the
> effect of resistance varying with temperature, power consumption by the
> lightbulb will be close to 1.65-1.72 times that at 120V rather than
> double.
> Keep in mind that a voltmeter will read a voltage other than about
> 169 volts unless it is a "True RMS" type. It may read 120 if it has
> fullwave rectification or possibly either 240 or close to zero due to
> having mere halfwave rectification. On a DC range, a non-true-RMS
> voltmeter will read halfwave rectified 240V or fullwave rectified 120V
> as about 108 volts - the actual average voltage. (Nominal voltage of AC
> is the RMS voltage, about 11% higher than the average for a sine wave.)
> According to what I consider a "usual 1-size-fits-all" rule, life
> expectancy is reduced by a factor of about 60 - and actual life
> expectancy results can vary significantly, even greatly.
> I would caution that burnout may be more spectacular than at 120V,
> possibly even unsafe, especially for a diode unless it is big enough to
> not protect your fuse or circuit breaker by blowing first (possibly
> explosively) should a "burnout arc" form.
> And at 70% overwattage, the bulb may overheat and break - especially in
> an enclosed fixture or a recessed ceiling fixture.
> Lower wattage 120V lightbulbs, such as maybe around or under 25 watts
> (?), may have their filament vary enough in temperature over each AC
> cycle when used with a diode and 240V to change results somewhat from
> results with actual 169-170V - probably for the worse in terms of life
> expectancy.
> - Don Klipstein (don@misty.com)
> ***********************************
> That seems to agree with Mark's explanation....
> Jeff
> --
> Jeffry Wisnia
> (W1BSV + Brass Rat '57 EE)
> The speed of light is 1.98*10^14 fathoms per fortnight.

Well, in that case, I lose my argument. life goes on, happy as before for
me. :-) I thought I was right, but I have been wrong before at least once.

Posted by Mark on November 18, 2008, 9:45 am

> Well, in that case, I lose my argument. life goes on, happy as before for
> me. :-) =A0I thought I was right, but I have been wrong before at least o=

nce.- Hide quoted text -

> - Show quoted text -

Hi Fartikus

try the other half of the experiment too...

compare the brightness of the bulb hooked to an actual 12VAC source vs
when it's hooked to the 24VAC + diode source. A lightbulb's
brightness will follow the true RMS value.

Unless your meter is a "true RMS meter" it will be lieing about the
rectified waveform reading. Most meters are not "true RMS" meters and
they read the correct RMS value only for a good sine wave which is
what we read most of the time.

Mark

Posted by Fartikus on November 18, 2008, 6:34 am

> Fartikus wrote:

Well, in that case, I lose my argument. life goes on, happy as before for
me. :-) I thought I was right, but I have been wrong before at least once.

Page 5 of 6 < 1 2 3 > last >>

Similar Threads	Posted
thermostat calling for heat, not blowing then thermostat switches to heat 2	October 24, 2008, 8:37 am
Looking for a new thermostat?	June 29, 2007, 1:41 pm
2nd thermostat	May 3, 2007, 2:54 pm
Need thermostat	September 30, 2008, 11:02 pm
Digital Thermostat	July 9, 2006, 12:26 pm
replacement thermostat	March 19, 2007, 7:54 am
Thermostat recommendations	July 11, 2007, 8:23 am
window AC thermostat out of cal	June 4, 2007, 8:54 am
thermostat bypass	June 8, 2007, 8:32 am
overiding a thermostat	June 25, 2007, 2:00 pm

Indicator Lamp on Weathertron 3AAT80B1A1 Thermostat - Page 5