Searching \ for '[EE] Ghost in the (AC) machine - Broken AC' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/power.htm?key=ac
Search entire site for: 'Ghost in the (AC) machine - Broken AC'.

Exact match. Not showing close matches.
PICList Thread
'[EE] Ghost in the (AC) machine - Broken AC'
2011\07\10@210411 by Herbert Graf

picon face
So, I've got an issue that I've pretty much run out of ideas on, and I
was wondering if a new set of eyes might help.

The central air unit at my home has stopped cooling at around noon, on
Sunday, the past 2 Sundays. It starts working again around 8-9pm.

The symptom is the unit is "on", i.e. the condenser fan is running
(including the furnace fan), but the compressor is not. The coolant
pipes are both ambient temp.

At around 8 or 9pm it starts working again, and works for the entire
week.

I've tried stopping and starting it and never observed the problem
outside this time window.

The things I've done/checked:

- the unit is receiving power and the thermostat is instructing the unit
to cool (the condenser fan runs, also I've measured 26VAC on the
solenoid terminals). I've pulled the panel and measured power on the
compressor terminals, meaning the solenoid is working (I knew that
anyway since the condenser fan is running). No current is flowing throw
the compressor leads when it isn't working.
- the starting cap measures in range (4.5uF for the fan, 34.5uF for the
compressor. Rating on the cap is 5uF and 35uF +/- 5%). Voltage present
on both running and starting terminals on the compressor.
- I've ohmed the wires in the unit, bending and stretching them to try
to isolate a marginal wire, nothing.
- I've gotten an R22 recharge pressure gauge, designed to connect to the
low side port. With the unit off I've measured somewhere around 200psi
on both low and high side ports (I can't give an exact figure, the gauge
I have is designed for charging, so it's linear up to 120psi, and then
goes non linear to 350psi, the needle was pretty much halfway between
120psi and 350psi). 200psi matches what I've looked up for R22 at an
ambient temp of about 100F (the high today was in the low to mid 90s).
- I've measured when everything is working perfectly, the low side port
is between 75 and 80psi.

So, that's pretty much all I've been able to come up with testing. As
far as I can tell everything is good, the bloody thing just won't start
it's compressor between 12 and 8pm on Sunday. (I have sensors on the
cooling lines and the air streams so I know remotely how the system is
doing, I see it cycle as usual, then nothing, just fan on, until late
evening).

The thermostat is a 7day programmable, the programming is:
Mon-Fri: 6am - 73F, 8am - 78F, 6pm - 73F, 10pm - 71F
Sat-Sun: 6am - 73F, 10pm - 71F

I can't really call anyone in since by the time they'd get here
everything would be working again, and I'd be out $150.

So, at this point I'm waiting for the problem to get worse, so I can
call someone in, but I was hoping that someone else here with any HVAC
experience might have any ideas? I still suspect the starting cap, alot
of online forums mention with these symptoms the issue is either low
coolant (my measurements suggest a correct charge) or a bad cap. I might
just get a cap and swap it in just to see if it helps things. Does that
make sense?

The unit is about 3 years old (house was built in 2008). It's a Fedders
condensing unit. Warranty is 5 years, but only covers parts, so labour
is up to me to cover.

Thanks to any suggestions anyone might have!

TTYL

2011\07\10@213716 by Carl Denk

flavicon
face
Does your power company control the air conditioner during peak electric times. Is there a control box that communicates with your power company. It might be a relay on the compressor power contactor  circuit. Normally this shutdown is for brief periods, maybe 10 minutes at a time like a rolling blackout. The compressor is the high power user. By just shutting down the compressor briefly, you probably would barely notice the event. The fan continues to operate. Could be a failure of the power company switch, or they are purposefully shutting down for an extended time period. Time for a conversation with the power company. It is likely the building owner some time agreed to have the controller installed, and received some financial benefit in return. My power company will credit my bill $100 for installing the switch, and rate reductions are possible.

I have a similar switch that I am in preliminary design stage, control my standby generator (runs in summer off our natural gas well - no cost), if the power company is going to use rolling blackouts, the house will switch to the generator automatically.

On 7/10/2011 9:04 PM, Herbert Graf wrote:
{Quote hidden}

>

2011\07\10@225846 by Herbert Graf

picon face
On Sun, 2011-07-10 at 21:37 -0400, Carl Denk wrote:
> Does your power company control the air conditioner during peak electric
> times. Is there a control box that communicates with your power company.
> It might be a relay on the compressor power contactor  circuit. Normally
> this shutdown is for brief periods, maybe 10 minutes at a time like a
> rolling blackout. The compressor is the high power user. By just
> shutting down the compressor briefly, you probably would barely notice
> the event. The fan continues to operate. Could be a failure of the power
> company switch, or they are purposefully shutting down for an extended
> time period. Time for a conversation with the power company. It is
> likely the building owner some time agreed to have the controller
> installed, and received some financial benefit in return. My power
> company will credit my bill $100 for installing the switch, and rate
> reductions are possible.
>
> I have a similar switch that I am in preliminary design stage, control
> my standby generator (runs in summer off our natural gas well - no
> cost), if the power company is going to use rolling blackouts, the house
> will switch to the generator automatically.

That's a good thought, but no, my utility used to offer that, but
discontinued it before I became a customer. Shame too, since the
thermostat they install is controllable remotely both my them and by the
customer. I had it at the place I used to live at, was really useful to
leave the cooling off all day and manually turn it on when I left work,
vs. a set time every day.

Thanks, TTYL

2011\07\11@071733 by Carl Denk

flavicon
face
Is that controller still in the circuit?? Use a voltmeter or some other means to check continuity of wires and assure that that compressor wires go only to the thermostat, and actual air conditioner controls. I would think the issue is with the 24 VAC control to the compressor's contactor (relay). If necessary run a new temporary pair of wires to the bare components needed. #18 or heavier wire should be adequate for the control wiring. Need to find EVERY end of EVERY wire on the control circuit. The timing of the problem sure doesn't point to your equipment, but something else controlling. The power company might be using that signal for other purposes today, and you still have the switch (controller). Again what does the power company say? Daughter on another power company that used to control electric hot water heater still has controller on wall. When the water heater was replaced a few years ago, the controller was removed from the circuit, but is still on her wall.

On 7/10/2011 10:58 PM, Herbert Graf wrote:
{Quote hidden}

>

2011\07\11@073452 by M.L.

flavicon
face
On Sun, Jul 10, 2011 at 9:04 PM, Herbert Graf <spam_OUThkgrafTakeThisOuTspamgmail.com> wrote:
> The central air unit at my home has stopped cooling at around noon, on
> Sunday, the past 2 Sundays. It starts working again around 8-9pm.
>
> The symptom is the unit is "on", i.e. the condenser fan is running
> (including the furnace fan), but the compressor is not. The coolant
> pipes are both ambient temp.
>
> At around 8 or 9pm it starts working again, and works for the entire
> week.

If it happens at exactly the same time for two weeks, you can suppose
that it's a logic/controller issue instead of a mechanical issue.
Because the compressor works for the whole week except for 8 hours, it
would stand to reason that it really does work just fine.

Try manually enabling the compressor using the contactor that switches
on the compressor. I'm not sure how it's arranged on these units but I
guess you would probably be able to cause it to close with your
24/26VAC signal.

-- Martin K

2011\07\11@101448 by Herbert Graf

picon face
On Mon, 2011-07-11 at 07:17 -0400, Carl Denk wrote:
> Is that controller still in the circuit?? Use a voltmeter or some other
> means to check continuity of wires and assure that that compressor wires
> go only to the thermostat, and actual air conditioner controls. I would
> think the issue is with the 24 VAC control to the compressor's contactor
> (relay). If necessary run a new temporary pair of wires to the bare
> components needed. #18 or heavier wire should be adequate for the
> control wiring. Need to find EVERY end of EVERY wire on the control
> circuit. The timing of the problem sure doesn't point to your equipment,
> but something else controlling. The power company might be using that
> signal for other purposes today, and you still have the switch
> (controller). Again what does the power company say? Daughter on another
> power company that used to control electric hot water heater still has
> controller on wall. When the water heater was replaced a few years ago,
> the controller was removed from the circuit, but is still on her wall.

I measured voltage at the actual compressor terminals (had to take most
of the unit apart to get there). The compressor has power.

Thanks, TTYL

2011\07\11@101846 by Herbert Graf

picon face
On Mon, 2011-07-11 at 07:34 -0400, M.L. wrote:
{Quote hidden}

During the problem period the compressor has power, as measured at the
compressor terminals. The only solenoid is receiving it's 24VAC, and the
condenser fan is running, proving the solenoid is working. There are
nothing but wires between the solenoid and the compressor housing.

I've been looking things up, and it just happens that the past 2 Sundays
have been the hottest weekend days of the summer, by a couple degrees
(high of 30C on the 3rd, and a high of around 32C yesterday).

It's looking like I've pretty much covered everything, damn.

Thanks, TTYL

2011\07\11@102024 by Sean Breheny

face picon face
Herbert,

I believe that you said that this problem happens for several hours on
Sunday. Does it happen absolutely every Sunday? Have you tried
power-cycling the AC unit during the time that the problem is being
experienced?

Sean


On Mon, Jul 11, 2011 at 10:18 AM, Herbert Graf <hkgrafspamKILLspamgmail.com> wrote:
{Quote hidden}

>

2011\07\11@102524 by Herbert Graf

picon face
On Mon, 2011-07-11 at 10:20 -0400, Sean Breheny wrote:
> Herbert,
>
> I believe that you said that this problem happens for several hours on
> Sunday. Does it happen absolutely every Sunday? Have you tried
> power-cycling the AC unit during the time that the problem is being
> experienced?

It has happened the past 2 sundays.

The AC unit has been power cycled numerous times during the period
(lately mostly because I've had to take the panel off to measure and
check things).

Thanks, TTYL

2011\07\11@103642 by RussellMc

face picon face
What is MEANT to stop it. eg

- Is there a thermostat somewhere that is responding incorrectly to
the high temperatures.

- Is there eg a fire shutdown sensor that decides that really hot days
are a house fire?

Put a fan heater (for circulation) in the relevant vicinity and heat
things to toasty too much. What happens?


Russel

2011\07\11@104233 by Carey Fisher

face picon face
And you're sure the outdoor coolant line is not freezing up?


On Mon, Jul 11, 2011 at 10:36 AM, RussellMc <.....apptechnzKILLspamspam.....gmail.com> wrote:

{Quote hidden}

>

2011\07\11@104724 by Herbert Graf

picon face
On Tue, 2011-07-12 at 02:36 +1200, RussellMc wrote:
> What is MEANT to stop it. eg
>
> - Is there a thermostat somewhere that is responding incorrectly to
> the high temperatures.
>
> - Is there eg a fire shutdown sensor that decides that really hot days
> are a house fire?

The compressor terminals have power, I measured right at the compressor
housing. Unless the compressor has some device in it that I don't know
of? FWIW I know that compressors often have overload switches that take
a few minutes to reset (I've waited hours). Some also have integrated
low pressure cutoff switches, but considering the low side is correct
when the compressor is running I don't believe that's the issue.

> Put a fan heater (for circulation) in the relevant vicinity and heat
> things to toasty too much. What happens?

I don't know what you mean by that? The condenser is outside on mounted
on the side of the house.
Thanks, TTYL

2011\07\11@105127 by Herbert Graf

picon face
On Mon, 2011-07-11 at 10:41 -0400, Carey Fisher wrote:
> And you're sure the outdoor coolant line is not freezing up?

Yes, while the unit is working normally there is condensation on it, but
no frost. The evaporator also is not freezing up. When not working the
line is at ambient.

Even if it were freezing up, my experience is it shouldn't take 8 hours
to recover, especially when it's in the upper 90s outside.

Thanks, TTYL

2011\07\11@105544 by Carl Denk

flavicon
face
That's my thinking, not mechanical, BUT, another fact that came out (not blaming anyone :) :) ), just happened to be hottest days on weekend. It could be likely that the compressor has a built-in thermal switch, and/or freon pressure switch. However I would think both of those would cycle in something less than an hour. If I am reading correctly, the power is being checked at the wires coming out of the hermetically (I'm assuming) sealed compressor, and the 24 VAC goes to a relay before that. There no other (ZERO) wires than the power ((2) 240 volt, ground, and maybe neutral) coming out of the sealed compressor?? Contact the compressor manufacturer and quiz them. They may even have a recall on the unit. Sorry about belaboring the very basics, but sometimes we can't see the forest due to the trees. :)

Get a contractor to replace the compressor with guarantee of results. :) A little expensive, but might be going there sooner or later. :~)

On 7/11/2011 7:34 AM, M.L. wrote:
{Quote hidden}

>

2011\07\11@110712 by Herbert Graf

picon face
On Mon, 2011-07-11 at 10:55 -0400, Carl Denk wrote:
> That's my thinking, not mechanical, BUT, another fact that came out (not
> blaming anyone :) :) ), just happened to be hottest days on weekend. It
> could be likely that the compressor has a built-in thermal switch,
> and/or freon pressure switch. However I would think both of those would
> cycle in something less than an hour. If I am reading correctly, the
> power is being checked at the wires coming out of the hermetically (I'm
> assuming) sealed compressor, and the 24 VAC goes to a relay before that.
> There no other (ZERO) wires than the power ((2) 240 volt, ground, and
> maybe neutral) coming out of the sealed compressor?? Contact the
> compressor manufacturer and quiz them. They may even have a recall on
> the unit. Sorry about belaboring the very basics, but sometimes we can't
> see the forest due to the trees. :)

There are only three wires attached to the posts on the compressor,
black, red and blue, black and blue go to the solenoid, red goes to the
starting cap.

> Get a contractor to replace the compressor with guarantee of results. :)
> A little expensive, but might be going there sooner or later. :~)

Unfortunately the compressor is under warranty. As such, I have to use
certain service people, and since the problem only happens rarely (so
far) all calling them will produce is a bill for $150 and a "nothing is
wrong sir" result.

Thanks, TTYL

2011\07\11@111439 by RussellMc

face picon face
> If I am reading correctly, the
> power is being checked at the wires coming out of the hermetically (I'm
> assuming) sealed compressor, and the 24 VAC goes to a relay before that.
> There no other (ZERO) wires than the power ((2) 240 volt, ground, and
> maybe neutral) coming out of the sealed compressor??

IF there is mains on the only two wires into the compressor, measured
with an additional external load across the terminals (say a light
bulb) and the compressor does not run
THEN  either
   it is broken internally
OR
     it has an internal switch that is designed to do this for some reason..
ENDQED.

I can't see how there can be any other options. (If there are then
Isaac Newton would have spotted them with ease :-) ).

I say "only two wires" to sidestep the 2 live + earth with break or
switch on one live wire and return path through ground scenarios*.

I mention the added load as this removes the prospect of there being a
high impedance interruption anywhere. Meter may see mains but there
may be high Z feed. This is not actually necessary for the above test
but adds unneeded belt and braces.

_________

* Long ago:

Appliances in general operated in house.
Appliances in general operated in sleepout.
Mains powered tape recorder worked in house
Mains powered taperecorder did no work in sleepout.

They asked me to investigate.
Transpired that sleepout had phase and neutral transposed (two live leads)
Recorder was wired from Phase to ground
                    123
Proper plug      PNE
Sleepout          NPE
Tape recorder  P_N

Appliances worked as PN is same as NP electrically.
Tape reorder worked as N is at E and vice versa so EN swap still runs.

BUT a number of appliances have N connected to E internally (despite rules)
Such appliances would work in sleepout but appliance body would be at
Phase = 230 VAC above ground.
Lack of a bench or basin or toilet in sleepout meant that nobody dies.

  Russel

2011\07\11@111518 by Carl Denk

flavicon
face


On 7/11/2011 10:50 AM, Herbert Graf wrote:
{Quote hidden}

Has anyone checked the refrigerant level (pressures)? Pressure switches could be set for both high and/or low pressures. Is the condenser clean - fins clear, fan blades clean, no bent fins, in general in good repair from an appearance viewpoint.  Fan spins freely, no bad bearings?
> Thanks, TTYL
>
>

2011\07\11@111650 by RussellMc

face picon face
> There are only three wires attached to the posts on the compressor,
> black, red and blue, black and blue go to the solenoid, red goes to the
> starting cap.

Rewrite my Sir Isaac Newton If ... THEN ... spec for the 3 wire compressor.
Feed it power as requiste and tick off options on chart.
Repeat with room heated with fan heater and a few hours allowed for it
to heat soak.



   Russel

2011\07\11@112010 by Herbert Graf

picon face
On Mon, 2011-07-11 at 11:15 -0400, Carl Denk wrote:
> Has anyone checked the refrigerant level (pressures)? Pressure switches
> could be set for both high and/or low pressures.
Yes, pressure both static and running seem to match what info I can find
online for a properly charged R22 system.

> Is the condenser clean
> - fins clear, fan blades clean, no bent fins, in general in good repair
> from an appearance viewpoint.  Fan spins freely, no bad bearings?

Yup.

Thanks, TTYL

2011\07\11@115123 by Carl Denk

flavicon
face

> Unfortunately the compressor is under warranty. As such, I have to use
> certain service people, and since the problem only happens rarely (so
> far) all calling them will produce is a bill for $150 and a "nothing is
> wrong sir" result.
>
> Thanks, TTYL
>
>    Have you discussed the situation with the service people, and tried to come to a low cost (both for them, and you) approach to the situation. Many times a team type effort works. Say if the unit goes down, you check the voltages to the compressor, and call them. They send someone out to verify your readings, that in fact the compressor is defective, under warranty. Then they repair on not overtime during week. If they come out on a Sunday, and the unit is defective, do you still have to pay. It seems that, the issue is isolated to a unit that is under warranty. Is this the same service people that did the install? Do they have a warranty, or just the compressor manufacturer's warranty? It would seem with the hot weather, they have service people on the road, and this would be a brief stop

2011\07\11@120739 by alan.b.pearce

face picon face
> Have you discussed the situation with the service people, and tried to
> come to a low cost (both for them, and you) approach to the situation.
> Many times a team type effort works. Say if the unit goes down, you
> check the voltages to the compressor, and call them. They send someone
> out to verify your readings, that in fact the compressor is defective,
> under warranty. Then they repair on not overtime during week. If they
> come out on a Sunday, and the unit is defective, do you still have to
> pay. It seems that, the issue is isolated to a unit that is under
> warranty. Is this the same service people that did the install? Do they
> have a warranty, or just the compressor manufacturer's warranty? It
> would seem with the hot weather, they have service people on the road,
> and this would be a brief stop.

It may even be a 'known problem' and they may replace the offending unit anyway ...
-- Scanned by iCritical.

2011\07\11@133920 by M.L.

flavicon
face
On Mon, Jul 11, 2011 at 11:10 AM, Herbert Graf <hkgrafspamspam_OUTgmail.com> wrote:
> Unfortunately the compressor is under warranty. As such, I have to use
> certain service people, and since the problem only happens rarely (so
> far) all calling them will produce is a bill for $150 and a "nothing is
> wrong sir" result.

This idea would make some people squirm: you could make sure there was
something wrong with the compressor. Merely a minor moral infraction
if you know the compressor is defective.

-- Martin K

2011\07\11@135520 by Herbert Graf

picon face
On Mon, 2011-07-11 at 11:51 -0400, Carl Denk wrote:
> > Unfortunately the compressor is under warranty. As such, I have to use
> > certain service people, and since the problem only happens rarely (so
> > far) all calling them will produce is a bill for $150 and a "nothing is
> > wrong sir" result.
> >
> > Thanks, TTYL
> >
> >    
> Have you discussed the situation with the service people, and tried to
> come to a low cost (both for them, and you) approach to the situation.
> Many times a team type effort works. Say if the unit goes down, you
> check the voltages to the compressor, and call them. They send someone
> out to verify your readings, that in fact the compressor is defective,
> under warranty.
The warranty is from the manu and only covers parts.

> Then they repair on not overtime during week. If they
> come out on a Sunday, and the unit is defective, do you still have to
> pay.
Yes. In fact, a service call during "normal" times, just to LOOK at
things, is expensive, coming out on a Sunday requires a "premium"
charge, didn't even bother asking what insane cost that would be.

> It seems that, the issue is isolated to a unit that is under
> warranty.
The unit is under limited warranty, it doesn't cover any labour or any
non covered materials, so if the unit were replaced I'd still have to
cover labour and recharge costs.

> Is this the same service people that did the install? Do they
> have a warranty, or just the compressor manufacturer's warranty? It
> would seem with the hot weather, they have service people on the road,
> and this would be a brief stop.

Brief stop or not doesn't matter in this industry from what I can tell.

TTYL

2011\07\11@141517 by Denny Esterline

picon face
On Mon, Jul 11, 2011 at 10:38 AM, M.L. <@spam@mKILLspamspamlkeng.net> wrote:

> On Mon, Jul 11, 2011 at 11:10 AM, Herbert Graf <KILLspamhkgrafKILLspamspamgmail.com> wrote:
> > Unfortunately the compressor is under warranty. As such, I have to use
> > certain service people, and since the problem only happens rarely (so
> > far) all calling them will produce is a bill for $150 and a "nothing is
> > wrong sir" result.
>
> This idea would make some people squirm: you could make sure there was
> something wrong with the compressor. Merely a minor moral infraction
> if you know the compressor is defective.
>
> --
> Martin K.
> --
>

I would never do something like that.
Certainly not if I bought a lemon of a laptop (with extended warranty) that
needed a motherboard replacement 16 days after I bought it (conveniently 2
days after the the store return policy). And certainly not if the
manufacturer serviced it another three times in 7 months before going out
of business. And I wouldn't dream of it after the extended warranty people
had it another four times before _they_ went out of business and was bought
out by a different company. And it would be unthinkable to do something like
that after sending it to the new company _three times in two weeks_ with a
complaint of "will not boot past POST" and getting back "works fine, could
not duplicate problem".

Yeah, after an experience like that, I'd never use jumper clips and hook
mains voltage to various places on the motherboard before sending it back a
fourth time. That would be morally questionable.

But it did come back with a new motherboard that time. :-)

True story (to the best of my recollection from 10+ years ago).

-Denn

2011\07\11@151626 by YES NOPE9

flavicon
face
Is the AC unit working today ? IF so,  can you heat the compressor above 100 degrees with a heat gun ?  Or a heating pad ?   Could we see some photos of the compressor ?  Is it possible to put an AC current probe on the wires to the compressor ?
99guspuppe

2011\07\11@164914 by Robert Rolf

picon face


YES NOPE9 wrote:

> Is the AC unit working today ? IF so,  can you heat the compressor above 100 degrees
 with a heat gun ?  Or a heating pad ?   Could we see some photos of the compressor ?
Is it possible to put an AC current probe on the wires to the compressor ?
> 99guspuppet

He would have better results running the compressor to get the windings hot. Insulate the motor with fibre glass to make it toasty like it would be on a hot day.

I had an old fan that died when hot. The issue was somewhere inside, the windings opened. The thermal cutout was ok.

I wonder if the thermal cutout/protector on the compressor motor is defective (age/heat lowering setpoint). It is usually buried inside the windings, but typically has a splice outside in the junction box that makes it accessible.
If you see AC across the cutout when the motor is off, you found your defect.

If you can show that the compressor windings have power, (put a light bulb across input terminals right AT the motor, but that it is NOT running, (video) then you can make your case for warranty replacement.

Good luck.

2011\07\11@165647 by RussellMc

face picon face
>> Is the AC unit working today ? IF so,  can you heat the compressor above 100 degrees
>  with a heat gun ?  Or a heating pad ?

Deja vu :-)

> If you can show that the compressor windings have power, (put a light
> bulb across input terminals right AT the motor

Deja vu :-)


Work ...


         R

2011\07\11@171213 by Herbert Graf

picon face
On Mon, 2011-07-11 at 13:16 -0600, YES NOPE9 wrote:

> Is the AC unit working today ? IF so,  can you heat the compressor above 100 degrees with a heat gun ?  Or a heating pad ?   Could we see some photos of the compressor ?  Is it possible to put an AC current probe on the wires to the compressor ?
> 99guspuppet


Yup,  working fine:


Note that it's set to go to 73F at 6am, and 78F at 8am. Also,
complicating things, the thermostat has a "recovery" feature, which
basically has it learning how the system responds, and pushes up
activation so that the temp set point is hit when the time actually
arrives. So, although I have it set to hit 73F at 6pm, it actually
turned on at ~3:20 so that it can bring the temp back down to 73F by the
time it hits 6pm:



Funny thing is today is the HOTTEST day we've had so far, hit something
like 34C (note the temp sensor isn't shielded complete from the sun, so
the early morning peak is an aberration), and everything seems to work
fine. Grrr...

2011\07\11@172510 by Herbert Graf

picon face
part 1 1273 bytes content-type:text/plain; charset="utf-8" (decoded base64)

On Mon, 2011-07-11 at 17:15 -0400, Herbert Graf wrote:
> On Mon, 2011-07-11 at 13:16 -0600, YES NOPE9 wrote:
>
> > Is the AC unit working today ? IF so,  can you heat the compressor above 100 degrees with a heat gun ?  Or a heating pad ?   Could we see some photos of the compressor ?  Is it possible to put an AC current probe on the wires to the compressor ?
> > 99guspuppet
>
>
> Yup,  working fine:
>
>
> Note that it's set to go to 73F at 6am, and 78F at 8am. Also,
> complicating things, the thermostat has a "recovery" feature, which
> basically has it learning how the system responds, and pushes up
> activation so that the temp set point is hit when the time actually
> arrives. So, although I have it set to hit 73F at 6pm, it actually
> turned on at ~3:20 so that it can bring the temp back down to 73F by the
> time it hits 6pm:
>
>
>
> Funny thing is today is the HOTTEST day we've had so far, hit something
> like 34C (note the temp sensor isn't shielded complete from the sun, so
> the early morning peak is an aberration), and everything seems to work
> fine. Grrr...

Ahh, nuts, didn't remember the list server strips inline images, my bad,
here are the graphs as attachments.




part 2 6482 bytes content-type:image/gif; name="1.gif" (decode)


part 3 6817 bytes content-type:image/gif; name="2.gif" (decode)


part 4 5869 bytes content-type:image/gif; name="3.gif" (decode)


part 5 181 bytes content-type:text/plain; name="ATT00001.txt" (decode)

2011\07\11@182205 by Robert Rolf

picon face

>>If you can show that the compressor windings have power, (put a light
>>bulb across input terminals right AT the motor

> Deja vu :-)

"Unfortunately the compressor is under warranty. As such, I have to use
certain service people, and since the problem only happens rarely (so
far) all calling them will produce is a bill for $150 and a "nothing is
wrong sir" result."

Hence the suggestion that he create a video recording that unequivocally shows that the compressor HAS power but that it is NOT running.
Hard to argue with a light bulb.

Sounds like his intermittent is temporarily fixed.
I would suggest he wire in the bulb anyway, so that when it next quits, he can record the status and make his case with the repair company that the compressor is intermittent since the light is on, but the compressor is not.
Want to bet they claim he voided the warranty by trying to diagnose the problem with a test light?


R

2011\07\11@182643 by Carl Denk

flavicon
face
Another thought, temporarily replace the programmable thermostat with a simple, heat/cool thermostat. I think it's only 5 color coded wires. Or even just connect the appropriate 2 wires for air. It still puzzles me, why only weekends. Have you talked to the manufacturer's tech. support. They usually will talk to the consumer, in particular if you are technically literate, which it sounds you are. Might be least expensive, to have someone else (another contractor) just replace the compressor. Then start over with a fresh warranty.

On 7/11/2011 1:30 PM, Herbert Graf wrote:
{Quote hidden}

>

2011\07\15@162408 by Barry Gershenfeld

picon face
Despite my finding this discussion late, I see we haven't passed another
Sunday yet.  My two cents has to do with the overload switch.  That switch
should operate if you stall the compressor. Note that things may have
changed since 1970 :)  We had a number of window units back then.  If you
stopped and restarted them without waiting, they would draw lots of current
until the overload switch opened.  Then there'd be a half-minute wait or so
before it tried again.

Modern thermostats, for sure, use their little 8-bit brains to prevent this
kind of abuse, so to do the test, you may have to operate some switches
manually (meaning, clip-leads).  Then, you can find out: Does the overload
switch work?  How long does it take to reset?  This is also a good way to
heat the compressor up.  At worst, you might push the poor thing over the
edge and break it completely.

Then, it's warranty work  >:-

2011\07\16@104243 by Justin Richards

face picon face
>
> The central air unit at my home has stopped cooling at around noon, on
> Sunday, the past 2 Sundays. It starts working again around 8-9pm.
>
If the times are exactly the same each time it fails then it must be
the controller and not the compressor so I suspect it fails "around"
noon on Sunday as you have said.

If it does happen again tomorrow at 12:00pm try setting its clock back
a few hours.

When you do a power cycle do you need to reset the clock, how does it
keeptime.  Does it have a battery backup to keep time.

It appears the controller is a wired controller and not a controller
via a remote control that is always pointed toward rhe reciever.

Is something near by in your street starting at around this time

2011\07\16@111037 by Carl Denk

flavicon
face
Just to say how obscure some of the event causes are: We have brief power outages due to birds or animals shorting out our circuit maybe twice a month (power company still troubleshooting the cause and remedy), but not every time there was an outage. When the power came back on (seconds, just long enough for the substation breaker to reset), our garage door would open. We noticed this about the time our smart meter was installed. But at about the same time, I was updating the whole house to modern "Decora" switches and outlets. As part of the process, I replaced the hard wired push button to operate the garage door, with a new one. I had not noticed that the new push button had a neon night light. After accusing the power company of opening my door, lots of research, I stumbled on the garage door opener's FAQ's. Buried there was a step by step eliminate the problem. Apparently the neon drew sufficient amperage to trip the garage door controller, but since the neon flickers and the power comes on somewhat random fashion, when both were synchronized some goofy way, the door would open. I opened the switch, removed the neon, easier than finding a new switch. :)

On 7/16/2011 10:42 AM, Justin Richards wrote:
>> The central air unit at my home has stopped cooling at around noon, on
>> Sunday, the past 2 Sundays. It starts working again around 8-9pm.
>>
>>      
> If the times are exactly the same each time it fails then it must be
> the controller and not the compressor so I suspect it fails "around"
> noon on Sunday as you have said.
>
> If it does happen again tomorrow at 12:00pm try setting its clock back
> a few hours.
>
> When you do a power cycle do you need to reset the clock, how does it
> keeptime.  Does it have a battery backup to keep time.
>
> It appears the controller is a wired controller and not a controller
> via a remote control that is always pointed toward rhe reciever.
>
> Is something near by in your street starting at around this time.
>

2011\07\16@111603 by Carl Denk

flavicon
face
Daughter has one of those fancy thermostats on a heat pump setup. One weekend she was having a family gathering on a hot day, and the air went down. What I remember similar symptoms. The service guy spent 2 hours, with wires loose everywhere, before he figured out the thermostat battery died after about 6 years. These fancy thermostats have been around just long enough that their batteries should be replaced as a routine service item.

On 7/16/2011 10:42 AM, Justin Richards wrote:
>> The central air unit at my home has stopped cooling at around noon, on
>> Sunday, the past 2 Sundays. It starts working again around 8-9pm.
>>
>>      
> If the times are exactly the same each time it fails then it must be
> the controller and not the compressor so I suspect it fails "around"
> noon on Sunday as you have said.
>
> If it does happen again tomorrow at 12:00pm try setting its clock back
> a few hours.
>
> When you do a power cycle do you need to reset the clock, how does it
> keeptime.  Does it have a battery backup to keep time.
>
> It appears the controller is a wired controller and not a controller
> via a remote control that is always pointed toward rhe reciever.
>
> Is something near by in your street starting at around this time.
>

2011\07\21@131556 by YES NOPE9

flavicon
face
I started watching *ghost in the shell* yesterday at a friends house....
I plan to return and finish watching it.

What was the ultimate ghost in your machine ?

99guspuppe

2011\07\21@140324 by Herbert Graf

picon face
part 1 3131 bytes content-type:text/plain; charset="utf-8" (decoded base64)

On Thu, 2011-07-21 at 11:15 -0600, YES NOPE9 wrote:
> I started watching *ghost in the shell* yesterday at a friends house....
> I plan to return and finish watching it.
>
> What was the ultimate ghost in your machine ?

I'm still "collecting data", I didn't want to bother the list till I had
enough data.

It hasn't actually happened "as bad" as the first 2 times.

Since I'm rarely home when strange stuff happens, I've added additional
sensors.

I now have proper gauges of both cold air temp in the furnace, and the
low side line coming out of the furnace. I also have a sensor on the
thermostat cooling signal, and I have a webcam trained on my power meter
monitor (so I can confirm when the compressor is actually running).

So, some stuff I've learned:

when it's "warm" outside (say about 75F), and the compressor hasn't run
for a while it takes about 5 minutes between when the thermostat tells
the AC to engage, and when the compressor actually starts up.

Unfortunately I don't know if this is normal behaviour? The last home AC
unit I had was about 20 years old, and the compressor started
immediately with the condenser fan started (which is when the thermostat
told it to cool). Perhaps high efficiency AC units act differently? Can
anybody comment on their experiences? It's a Fedders outdoor unit, I
believe 1.5ton, operating on R22. Can anybody on the list with a similar
unit comment on when there is any delay between the condenser fan
turning on, and the compressor turning on?

When it's REALLY hot, things are different. I haven't noted an exact
correspondence, but this is the "feel" I get:

- if it's really hot (say in the 90s) and the thing hasn't run for a
while, it take about 5 to 10 minutes for the compressor to start
- if the thermostat is cycling at a set point, the compressor will start
within about 5 minutes the first time, more then 10 minutes the second
time, and up to an hour the third time
- once the compressor is running it will run for HOURS without issue,
it's the cycling the seems to be the issue.

I think the reason I only noticed the problems on weekends is due to how
I have the thermostat set:

- on weekdays I let it go up to 78 during the day and then at 6pm I have
it set to 73. On warm, but not hot days, there is pretty much no
cycling, by the time recovery mode kicks in on the thermostat (usually
around 3 or 4pm) the house hadn't yet reached 78F
- on hot days the house reaches 78F before the recovery mode triggers,
and I see the 1 hour start time after a few cycles
- on the weekends I have it set to 73 pretty much during the whole day,
so it's always cycling, that's why I think the problem was more
prevalent on the weekends

I'm still kinda at a loss as to what to do. I'm going to call some AC
people in the area and see if this situation sounds familiar to them. I
just don't want to get ripped off, so I'd much rather have someone tell
me what's going on who DOESN'T have an incentive to rip me off! :)

Thanks for any comments/insight/critisism!

TTYL


part 2 6195 bytes content-type:image/gif; name="1.gif" (decode)


part 3 181 bytes content-type:text/plain; name="ATT00001.txt"
(decoded base64)

--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
mailman.mit.edu/mailman/listinfo/piclist

2011\07\21@151745 by RussellMc

face picon face
> I'm still kinda at a loss as to what to do. I'm going to call some AC
> people in the area and see if this situation sounds familiar to them. I
> just don't want to get ripped off, so I'd much rather have someone tell
> me what's going on who DOESN'T have an incentive to rip me off! :)

As a last resort, if you can't get reasoned input elsewhere - and you
probably can - I know a very practical and experienced air con man in
Canada who would probably be able to comment usefully.



                Russel

2011\07\21@160859 by jim

flavicon
face

Herbert,

Just a thought I had while reading this.  If for some reason when the
compressor turns off at the end of
one cycle, and then the thermostat calls for more cooling and the
pressure hasn't beld down on the
compressor, you may have a condition where the compressor is trying to
start while there is still a
pressure head on the high side of the line.  If this is the case, the
compressor either won't start at all,
or will struggle to start, and wind up trippin the overheat sensor and
cutting power to the compressor.  If
that happens, you'll have to wait for the heat to subside before the
thermal switch can reset.  This might
take anywhere from a few minutes (warm outside - 75F) to maybe an hour
or moe (hot - 90F+)
 I don't know if this is what's happening, but the symptoms sort of
support the theory.  Maybe it's
something you could watch for in your investigation with all of the
sensors you have available.

Just food for thought.
 Regards,

Jim

> ---{Original Message removed}

2011\07\21@162028 by Herbert Graf

picon face
On Thu, 2011-07-21 at 13:08 -0700, RemoveMEjimTakeThisOuTspamjpes.com wrote:
{Quote hidden}

That is exactly the train of thought I've been going on. The thermostat
seems to have a 10 minute guard period, after it turns off it doesn't
ever seem to turn on again until 10 minutes have passed. I'm wondering
if that's long enough.

Something I didn't want to mention since it's so fleeting is every once
in a while, when I see the "long time to start" situation, my power
meter registers 2 to 3x normal power usage for the 1 minute interval
(normal power usage is 1.2kW, I've seen 3 and 5kW reported).

If this is the case, perhaps I can either set the thermostat to have a
longer guard interval, or replace it with one that has a longer one.
Does 10 minutes sound right to people?

The good side of all this is I've learned ALOT about AC systems, enough
that I might experiment a little when I have time.

Thanks, TTYL

2011\07\21@163649 by Bob Ammerman

flavicon
face
> Something I didn't want to mention since it's so fleeting is every once
> in a while, when I see the "long time to start" situation, my power
> meter registers 2 to 3x normal power usage for the 1 minute interval
> (normal power usage is 1.2kW, I've seen 3 and 5kW reported).

This could be a clue that the compressor is stalled fighting a head on the high side, as the current hyposthesis states.

-- Bob Ammerman
RAm Systems

2011\07\21@171122 by YES NOPE9

flavicon
face
Hard Starting AC compressor
appliancejunk.com/forums/index.php?topic=3290.0
     Only a few minutes should be required to drop the high side pressure.  Longer than that indicates a problem I should think.

I recommend putting a pressure sensor or gauge on the pressure high side of the system.......   perhaps there is a stub on the high side.
You could always put a camera snapping photos every 30 seconds on the gauge..   It is easy to get gauges with automotive recharge kits.

Have you posted photos of your AC system ?  
99guspuppe

2011\07\21@184355 by RussellMc

face picon face
> Something I didn't want to mention since it's so fleeting is every once
> in a while, when I see the "long time to start" situation, my power
> meter registers 2 to 3x normal power usage for the 1 minute interval
> (normal power usage is 1.2kW, I've seen 3 and 5kW reported).

Didn't feel like giving us too big a clue then ? :-)

Theories:

1) How about - compressor trys to start early, pressure too high.
Won't start into load so just powers one phase or single phase or
whatever.
3 to 5 kW!
That baby is tough !
Too many more hot Sundays and you'll have  a warranty call out yet.

or

2) Something else ... :-)


More... (looser matching)
- Last day of these posts
- In 2011 , 2012 only
- Today
- New search...