DIY 230vac adjustable power limiter
 

This 230vac adjustable power limiter can be pre-set to turn off AC power
to a load and reset itself within a few hundred milliseconds.

http://i46.photobucket.com/albums/f1...NCL/AB2011.jpg

The blue device on the left, is a non-contact current sensor.
It's set-point has been set to about 9.5 amps. (~2300 watts).

The dark blue box (lower right) contains a transistor switch that sends
5 VDC to the two SSR on each side of the (open) 10A breaker in the middle.
(The green LEDs on the SSRs indicate they are on).


If the load exceeds 2.3kw, the current sensor closes it's contacts.
This grounds out the base of the transistor, shutting it off.
That removes the +5v control signal from the SSRs, shutting them down.
(The SSRs turn off as the AC voltage crosses Zero volts).

~~~ The Sanyo hiccups! :D

Once the current stops, the current sensor shuts down.
(It gets it's power from the magnetic field of the 230vac line).
It's contacts become open. A cap on the base of the transistor
charges up and the transistor turns back on.
That turns on the SSRs and the Sanyo restarts within a few seconds.
(The SSRs turn ON as the AC voltage crosses Zero volts).

Pretty slick little setup. :)

What happens when the Sanyo restarts?

Heat sinks is staying below 70F
 

The first test was done with High Power mode on.
Which is always good for a big ramp up in power use. Maybe around 2.6 to 3kw..
When it hit the set-point, it went quite for a second, started right back up,
but at normal power levels (460w).
Then started the stepping-up sequence again, heading for 3kw,
but never getting there. At 2.3kw, it hiccuped again.
After three hic-cycles, I took it out of HP mode.

Right now, we have a lot of snow and ice on the coil, so I'm waiting to see
if we get a defrost cycle. That would be a good test of the unit.

If it can get into defrost mode (where it normally uses 1400-1600 watts),
I'll be very interested in seeing what happens when the famous
After-Defrost-Surge* automatically kicks in.

If the auto-breaker's short drop-out can make the Sanyo restart and run
up to a more reasonable power level, I'll be happy.

If it can't save itself from repeating the defrost power surge after
the drop-out, then I'll change the time delay and made the delay a few seconds.
That will insure a normal power-up cycle without a crazy surge.


~~~
* After-Defrost-Surge note:
My theory is, that some Japanese management guy decided that Americans
are so impatient, they will want a big strong blast of heat, within seconds,
after the defrost cycle.. So, set motors for ludicrous speed!

It worked!! Read the latest news over at
http://ecorenovator.org/forum/geothe...html#post12588

Little blue box
 

In the lower right side of this picture, there is a little blue box..

http://i46.photobucket.com/albums/f1...NCL/AB2011.jpg

It contains this little transistor circuit.

http://i46.photobucket.com/albums/f1...Solar/ACBQ.jpg

The AC current sensor closes it's contacts at ~9.5A and shorts out the voltage
from R2, that's been keeping Q1 turned on. Once Q1 turns off, the SSRs
also turn off, causing the Sanyo to Hiccup. :D

~~~

Last night and this morning, we had a little snow storm. (it's still snowing now)!
The wind was blowing the snow around, the temperature was 32F, the Dewpoint was 31F and the humidity was 99%..

Snow and ice was quickly covering up the Sanyo air intake.
This AM, we started watching the defrost cycles. Smooth as black ice!
This is the way my Sanyo was meant to work! :)

Hacking Sanyo outdoor units?
 

http://i46.photobucket.com/albums/f1...CL/outdoor.jpg

Looking at this diagram of the Sanyo Outdoor Unit, I can see how the
Power Relay (upper right) could be easily replaced with a 240vac SSR.

I would bring in 5VDC from outside to control the SSR, keeping the High
Pressure Switch(HPS) and OverLoad Relay (OLR) in series with the 5VDC.

It would not be difficult to add a current sensor(@9.5A) with
it's control circuit (see above) in series with the OLR & HPS contacts.
(The transistor would control the ground side of the 5v power source).

Thus eliminating the need for that messy looking relay rig inside the house.. :)

During mild weather, the crankcase heater and other vampire loads could
be dropped off the grid, by using a spare X10 to control the indoor 5V power supply.

~~~

It just occurred to me, a second Sanyo Outdoor unit (@9.5A) could
possibly be connected to the same outdoor 230vac power disconnect box..
Assuming 20A breakers were installed in mains box.
Saving a few bucks for another permit.. :p

Power control hack
 

Here's the before pic.
http://i46.photobucket.com/albums/f1...CL/outdoor.jpg

This is going to be the after Hack pic..
[IMG]http://i46.photobucket.com/albums/f1...L/Untitled.jpg[/IMG]

Terminal Plate Connector (5) is the switched side of the 230 supply.
The SSR connects (5) to (1), which feeds the indoor unit and the main PCB ACIN1.
The Transistor block will be the same as used above.
The DC power supply puts out 12VDC with AC inputs from 100 to 240 50/60hz.
Current sensor will be pre-set to drop power at 9.5 Amps.. (~2.3 kW).

This might void the warranty, so I am considering installing it outside the unit. (As above).

One added benefit of using an SSR is no contact spark.
Safer if there's any Propane leaking in the area. (BBQ tank on the back deck). :rolleyes:

This looked much neater in the diagram!
 

http://i46.photobucket.com/albums/f1...CL/breaker.jpg

Three Wires:
brown wire from the current sense loop goes to (1)
brown wire from the SSR goes to (5)
White wire with spade lug goes to the WHT wire (high pressure sw etc).

This will be installed on the repaired (non-warranty) unit tomorrow.
Hopefully, those Ebay parts will do the job.. :D

The Transistor box is from a recycled garage door Open wireless sensor.

Auto-breaker installed
 

It looked so neat before the hack.. (The compressor compartment is under this plate)
http://i46.photobucket.com/albums/f1...L/BeforeBH.jpg

Afterwards, it has a kind of Hacked look..
http://i46.photobucket.com/albums/f1...CL/afterBH.jpg
No wires were cut and only one hole was drilled to bolt down the SSR.
So, if anything fails, the old relay can be reinstalled in about 5 minutes.

The current sensor is set for 10.0 amps and seems to be working perfectly.
The system (indoor & outdoor units) should never draw more than 2.4 kW.
Once the sensor trips, the indoor unit will reset and start where it left off,
before the power outage.

I am gratified that I found your post. It prompted me to join this forum. I am considering the use of SSRs to add more control to my enphase micro inverters. The original thread is on the wind-sun forum under the heading: 10.78kW Enphase system and XW6048 (ForumVB/showthread.php?t=12041)

Thank you for making your information available. I plan to use it!!

Rex

Hi Rex,
I just looked at some of your recent posts over on wind-sun. Wow, that's a big system!

Anyways, SSRs from China work the same as the SSRs I've used at work.
Bolt them down on some metal and they run pretty cool. I've never had one fail.

The current sensors aren't cheap, but you can find them on Ebay.
VERIS H735 CURRENT SENSOR/ RELAY HAWKEYE 735 | eBay

I just picked up a spare current sensor last week. :)

Have you seen IOTA auto-change-over relays??
http://ecorenovator.org/forum/solar-...er-switch.html
I'm using some and they work nice. But, they are a bit slow when De-energizing.

I think a 240vac line can be switched much faster using four SSRs.
But, the DC control signals into the SSRs have to be very well controlled, and super reliable.
(To avoid a short). :(

I'm switching just one line of 240vac in my Sanyo #2 since it does not use any 120.. Just 240.
Two SSRs weren't necessary, to mimic the Sanyo power control relay setup.


~~~
I really like those Enphase inverters and I'm thinking of buying one or two,
M210 units to experiment with.. ;)

Cheers,
Rich

I bought the largest that would fit on my south facing roof. It's an all electric house and I have an electric car. So I'll probably still end up paying for some electricity, but the gas money saved will easily cover it.

I have explored several ways of controlling the enphase inverters and SSRs seem to be the best. I finally bit the bullet and had the PV installer run 11 strings of 4 panels to an ac combiner panel. That will allow me to use AC SSRs more easily and avoids the DC side of the PV panels completely.

I have plans to AC couple the micro inverters. The combination of the current sensor switch and the SSRs may work. Still waiting for the electric company to commission the system before I get to ambitious making changes.

You are way ahead me! I looked, briefly at ABT switches and was put off by the cost. These beauties are sending me back to re-evaluate my plans.

Thank you again!
Rex

Your projects are so ambitious! I gave up on my dream to get a hybrid.
And you went to a Full EV! Wow! That kinda move takes some guts.
(I still have an electric bike on my bucket list). ;)

The problem with my IOTA switches is, they are 120vac, so I can't use them to
take my heat-pumps off the grid and use the solar backup system.
Just have to make sure I use UL listed SSRs.. :)

Cheers,
Rich

I predict your next vehicle will be all electric. My wife and I have second car that is still gas :(

Before I start designing my control system, a question.
You said you are controlling the Sanyo 240VAC with an SSR in one leg of the 240VAC. Is that ok? Because that would lower the cost of controlling the 11 strings by 1/2!!

Regarding the IOTA switches, one leg of 240VAC is only 120VAC and normal 120VAC is 120VAC. So can you use 2 IOTA switches, one per leg? Or can you use one IOTA in one leg like you do with the SSR for Sanyo?

My EV is destined to be a bicycle, I'm thinking fount hub..

On the first auto-breaker, I used dual SSRs to completely disconnect the 240 from the Sanyo.
The hardware is all indoors.

But, when I wanted to install SSR control inside the #2 Sanyo system,
I saw they were only shutting off one side of the line.
Since they don't use 120 inside the unit (like my 240 laundry dryer), they don't need to use the Neutral line.
The unit is powered by two wires on terminal connections 5 & 6. The chassis ground is for safety.

~~~~
Since your Enphase power sources are need to be able to feed both 240 & 120 loads in your home, the Neutral line is a big deal.
And since you have it, you will need to disconnect Both sides of the 240 hots.

If you lived where they used only 240 loads in the house.. Maybe one side would work okay..

~~~

It seems like it would be easy to convert a 120 IODA to switch 240.
Just pull one side of the relay coil driver/timer off the line and connect it
to the Neutral. That would give the coil 120.. And it wouldn't melt.. :D

You have a good DIY sense, hopefully you can help me lower the cost of my PV control system.

The goal is to have an automated control system that will:
1. When grid power is up:
1.a. All strings of PV panels connected to grid
2. When grid power is down and sun is up:
2.a. Measure house power consumption
2.b. Measure PV power production
2.c. Match the a & b by using:
2.c.1) adjustment of diversion load
2.c.2) turn on/off a string of PV panels and adjusting diversion load

A string of PV panels consists of 4 - M215 inverters with 244watt panels which should produce 860watts at 240vac.

I have looked at using SSRs with a commercial controller, it should work and it is expensive. That would be on top of the islanding inverter & batteries still needing to be purchased.

What is the diversion load going to be?
Tell me more about the 'islanding inverter' too please..

Diversion load? Haven't decided yet. I think I need something variable upto about 1500watts. A hair dryer or something might work. That would balance almost 2 PV strings of power.

The islanding inverter I want use is the XW6048. Installation manual http://www.global-download.schneider-electric.com/852577A4005D7372/all/9A19F733940214AA8525781F0070D57C/$File/xw-hybrid-inverter-charger_installation-guide(975-0239-01-01_rev-e).pdf
I think it will be big enough to supply my normal house loads. Especially if I can talk the wife into using the mini split A/C systems. Gonna be tough 'cause "they are ugly"! If it saves enough power/money I think it looks good, very good. The Leaf is gorgeous!!

The details aren't firm yet due to lack of power consumption data. I have Brultech energy monitoring system that I had installed prior to the PV system install. It got removed during the service entrance upgrade and I don't want to reinstall it until after the PV system has been commissioned. After I get enough info to make a good decision, I will move forward with the purchases.

There is a new inverter that just came out that I am evaluating as well. The Outback Radian GS8048 OutBack Power / Products / Sinewave Inverter / Radian

Looks like you've got the Island hardware covered. (That's a big manual)!

If you want super efficiency heating and cooling (and don't live too far north),
use Inverter type Mini-split systems..

Sanyo #1
http://i46.photobucket.com/albums/f1.../NCL/Sanyo.jpg

Sanyo #2
http://i46.photobucket.com/albums/f1...t/downpout.jpg

Give you an idea of how big these things are.. I'm 6' tall..
http://i46.photobucket.com/albums/f1.../10goofing.jpg


When we put #1 in, my wife had a LOT of reservations about how it would look.
But, she was super pleased by the system's performance.
Pretty soon, 'looks' weren't so important.

But, when I put in #2, the lines were out in the open. She didn't like the idea,
but the price of installing this unit was pretty low..

And the outdoor unit was an old repaired unit, just taking up space in the garage.
(On her side of the garage)! :D

It's been super warm in the house so far this winter.
The 2 systems together add up to 48,000 BTUh or 4 tons of cooling.
Under idea conditions, the max heating is 72k or 6 tons!

When they are running, about 460 watts from each unit will maintain the set-point (on the average winter day).
Unless it's really cold out, they will coast down to standby wattage (30 to 60w) after setpoint is hit.

So, if it's pretty cold, with no solar gain at all, and they both ran 24-7, using 1kw total,
24 kwh will run us about $5.19 cents a day or $155.70 per month. (My best guess)..

Which isn't too bad, unless you can get heating oil for a buck a gallon.. :p

Since it's mild here in the summer, (compared to Texas) cooling cost us
much less. I figure August will run less than 5 kwh a day (average).. Peanuts.

Hey, I just remembered.. LQ has some neat looking picture-frame style IDUs..

Cheers,
Rich

I live in AZ so I'm hoping the COOLING capacity will be enough. I was thinking of adding a mini split (#1) to the master bedroom to steady the temperatures. Right now the main A/C kicks on & off often enough to be bothersome. I plan to use an inverter mini split (I thought all mini splits were inverter type mini splits) to lower the cost of cooling the room. I might have to put it in and see if she notices...

The Islanding inverter will work, the issue is making the micro inverters work with it. I think I'm on the right track with using the SSRs and and Arduino controller. But I have not worked with an Arduino before, so there is the learning curve to climb. But I have to admit the DIY aspect is very enticing.

After #1 mini split is running comes #2...

Our cooling is awesome. One unit cooled the whole house this past summer.
We did have a few hot spells. But, a few days in the 90s is nothing compared to AZ..

You can get older tech non-inverter units.. They are cheaper and less efficient.
Low SEER, with fixed speed compressors. Nothing you want.

Some modern hi SEER units can run in a 'Quiet' mode. Or in low fan..
That allows them to run very quietly.. Most of the time, you can't tell when the are on.
Even the outdoor units are low noise..


Here's the only good pic of a remote I could find on-line.. Mine looks much better..(More buttons).. ;)
http://acsuperstore.com/Merchant2/gr...UnitRemote.jpg

I was thinking the Sanyo 2 ton unit would be more than enough to cool the master bedroom & bath. Most likely I'll breakdown and get the software to do a heat load calculation. Maybe I'll get lucky and the mini splits will get even better.

If it's 35'x40' two tons should work..
Depending on the amount of insulation and glass.


http://www.youthmagz.com/gambar/room...zing-chart.gif


SEER makes a difference too..
http://realpages.com/sites/sheltonsh...s/pg3chart.jpg

The room is not that big, it looks like a one ton might work.

But if you include the bathroom area, which is open to the master, then the larger unit might be needed. Have to do some measuring! Plus the bath has a glass block wall, which is pretty, that does a poor insulating job compared to a regular window.

My large room is the Den, 18x16=288 sqft
http://i46.photobucket.com/albums/f1.../9784_9784.jpg

So, a 7k BTUhs would likely do the cooling. Due the 100 sqft of glass (none of it southfacing),
I think 12k-14k is needed for heating..
I wanted to buy a small 9-10k mini-split.. But my wife was okay with cleaning
out the garage and using the old 24k in the den.. So, it's about twice what we need.
The heating is super fast (when it's in the 30s) and I figure the cooling will be ultra-fast.

We had a light layer of ice on the ODU coils this morning, so I did a manual defrost.
Put both units into cooling mode for about 4 minutes. The IDU coils got down to the 20s real fast.
They quickly sucked enough heat outdoors to completely clean up both units..
Switched them back to heat mode and watched them both start running more efficiently.

I'm going to try to get a low temp/dew-point alarm system going.
So, it can tell me when things are icing up. Otherwise I am going
to need come video cams out there..

It seems there is some value to "over sizing" the mini split heat pump. How much "extra" does it cost to run an over sized, say 3 ton vs 1.5 ton unit. The enjoyment of faster temperature recovery cannot be directly measured, but does have significant value to those who are always too cold or too hot.

I'm not sure. I've only used a couple of 2 ton units.
One thing that I have noticed in the den, is a pretty short warming cycle, when it's mild out.
Yesterday and this afternoon, I noticed the ODU fan only ran for 30-50 seconds before shutting down.
On the TED power monitor, it shows 30-60 watts, then creeps up to 450-480
for 30-50 seconds and then back to 30-60 watts again..

The remote's heat sensor is enabled, and the remote only sends it's data once every 5 minutes.
So, when it's mild, I guess the system must be using it's IDU heat sensor to
keep it from over-heating the room. :confused:

Or, maybe an increase in heat to the remote, will trigger it to send a data bust earlier than normal.?.


When the ODU shuts down, the IDU fan stays on a while, since the coil is still warm.
After the coil cools down some it switches to super-slow-flow.
It uses super slow to slowly move air past the built in thermistor.
It's also slowly filtering the air in the room. :)

Sanyo#1 is heating a much larger space, and runs longer, when it comes on.
Maybe 4 or 5 minutes on mild days. But it might stay off longer too.

~~

Anyways, the Sanyo can throttle WAY back and work like a little 4k BTU wall AC,
but not use that much power..(300-480 watts).

http://i46.photobucket.com/albums/f1...L/poweruse.jpg
So, if you do get one too big, you should not end up using excessive power.

You should look at the specs of the units.. I think there might be a low end,
where you can't use any less power and still be pumping refrigerant. :p

I'm amazed to see my both units cranking out some good heat in the morning, running at 900 watt total.
Then it gets warm and they slow down.. 60watts total in idle mode..
(Heaters on the crankcases).

Scraping the transistor for a 555 timer chip..
 

After months of careful deliberations, I've decided to change my design.

I will connect the current sensor to (pin2) trigger the 555 timer.

Over-Current:
The output (Pin 3) of the 555 (connected to the cathode of the SSR dc control) will go high,
turning OFF the SSR for a controlled time period.
I was thinking of selecting C1 & R1 for a 5 minute time delay.

http://ecelab.com/circuit-monostable-555.jpg
http://www.kpsec.freeuk.com/images/555monop.gif

I want to make this change, in case there is a failure in the Sanyo controller,
that causes it not to reset, but to go nuts drawing excessive power..

In that case, a timer would insure there was no relay chatter or
Low frequency 230vac Buzzz! Which, could generate a lot of heat..

Defrost Speed Control:
A 5 minute cooling off period, combined with the 5 minute delay built into the remote,
might slow down a Defrost cycle, but should allow it to be done using less power.

Comments?

Xringer, the point of these is to stop the units from drawing excessive current during defrost or temp change calls from the thermostat right? What does Sanyo say about these spikes? Are they over the rated electrical consumption of the Minisplits? I feel like somewhere I've read that inverter driven units can run up to 120% as a "Boost" if need be, could that be what your units are doing, and thus totally normal/unharmful? Does this actually use more power than if it were to run at a normal load for a longer amount of time since it would achieve setpoint sooner and go into idle mode sooner?

Just trying to gain a little understanding because I find this interesting,
Adam

Yeah, I'm limiting the current to around 10A (around 2.4 kW) since the spike
wants to run up over 3.5 kW. (about 15A).

The last time I inquired, Sanyo USA didn't seem to know what do to.
Their expert told me to have the charge pulled and weighed.

They might even think it's okay to pull that kind of power..
Supposedly, there is already protection in the unit, to limit power to 5 kW..
http://i46.photobucket.com/albums/f1...er/NCL/5kw.jpg
(Why use 22.5A, when the max breaker they list is 20A)???

I obtained a 15A breaker, but the dang thing wouldn't trip..
I guess regular electromechanical breakers need a big overload to trip.?.
They rate them, so they can take the huge Start current of the old tech HVAC systems..


That seems very odd, since the Compressor Locked Rotor is 17.5 amps.
http://i46.photobucket.com/albums/f1...r/NCL/F106.jpg

As you can see, the system is rated for a max of 13.5A and that's just a tad over 3 kW..
So, IMHO going over 3.5 kW isn't where it should be operating.

Is it harmful? I'm not sure. But, I wouldn't want to try running my Ford Escape at 150 MPH for very long.. :rolleyes:

I suspect that very high pressure occurs in the R410A output line of the compressor, when power use exceeds 3.5 kW..
I also suspect that high pressure blew a hole in the output line, on my original ODU.
http://ecorenovator.org/forum/geothe...html#post13279


I have a few theories about why this problem is occurring.
#1 The control firmware is NG. Programming mistake.

Maybe the marketing guys told the engineers to add a big boost
of heat for the North American market, since Americans must have
instant heat, or they won't buy a Sanyo..
So, it was hastily done. The Big Boost function wasn't fully tested.
Maybe they only had 208VAC in their lab at the time??
Maybe their test unit didn't self-destruct? (Tested for 2 hours)?


I've just learned to accept the system, warts and all.
The work-around ,was to carefully watch power use and hit the off button,
when power started shooting up to 3.5 kw..

Adding an automatic power limiter means I don't have to get up at 3AM
to press the off button.. :)

53 seconds
 

220 k resistor and a 220 uF cap will give us a 53 sec delay.

220 x 0.220 = 48.4 & 48.4 x 1.1 = 53.25 sec


http://i46.photobucket.com/albums/f1...lot/mono-1.jpg



Pin 2, will be open (or high) unless the over-current sensor switch closes. Then it will ground to Zero volts.
Pin 3, normally low, will keep the SSR(s) turned on. (Cathode connection).
During an over-current event, the timer starts and pin 3 goes high, turning off the SSR(s).

http://www.kpsec.freeuk.com/images/555monop.gif

I think 53 seconds is a bit short, but I decided that a longer pause might
be too slow, and drag out the defrost cycle..

Right now, my main concern is to get rid of the old transistor circuit,
and it's very short delay time.
I worry that maybe the controller won't re-set every time the power just hiccups.
A 53 second hiccup seems like a more robust reset period.. :D


Edit: 1/13/2012 8:47PM
Maybe I should have tested my 220 uF cap.?. The time delay is 72 seconds.
Luckily, it's not critical. 72 seconds will work just as good.
The important thing is, the new board is built, installed & working perfectly.. :)

I plan to upgrade Sanyo#2's current limiter, but it's installed inside the ODU.
So, I'll have to wait for some nicer weather before changing it's circuit board. :(

Edit: 1/18/2012 18:50
The other day, it was near freezing and was damp outside.. As I walked into the living
room, I looked up and noticed the vent was closed on Sanyo #1.
The Remote control was still ON (in heat mode). Since the vent was closed,
I assumed a post-defrost current surge exceeded 10A and shut down #1 for 72 sec..

After 72 seconds, power returned and the system cycled into standby/off with it's vent closed..
I waited a while for the remote to transmit it's stream (Heat Mode, 21C, ION off & etc)
and Sanyo #1 came back up, operating normally..

I LIKE IT!! :thumbup:

555 timer for ODU #2
 

http://i46.photobucket.com/albums/f1...ot/555ODU2.jpg http://i46.photobucket.com/albums/f1...t/555odu2b.jpg

Using the same value parts, this one times out at 63 seconds.?.

It seemed to be triggering before I could put the ground in on pin 2,
but then I realized that I'm packing kV+ volts of static electricity today. :eek:
After self-grounding, the unit tests out okay for installation..

Now, I just need some nice global warming in the back yard for about 20 minutes.. :p

http://i46.photobucket.com/albums/f1...L/Untitled.jpg

Wow! global warming!
 

For my birthday, I got some good weather!!

High Temperature 54.5°F
Low Temperature 40.6°F

The new 555 timer unit was almost a drop-in replacement.
I did bring the 240vac-to-5Vdc power supply inside,
to add on some heavier wire for connecting to the terminals.
Got to use my new soldering iron (B-Day gift from wife).

The 555 is driving the SSR just fine so far.
Maybe we'll get to see that 63 second delay working,
if the lows get back down in the 20s next week..
(Causing defrost cycles).

The remainder of this month looks pretty mild..
Woburn, MA 01801 Month Weather Report - AccuWeather.com

Happy birthday! :)

We've had two over-current events this season so far. Not caused by a defrost cycle.

1. Wife turned up the heat from Sanyo #1 and it ran up to a peak of abt 2.5 kW (10A) and reset (with time delay).
She only called for 1 deg C increase.

2. A couple weeks later, I did the same thing with Sanyo #2. Reset for no apparent reason.


Both times, these events occurred in the early morning hours, while bringing
the temperature up from 20 to 21 C.
The over-night temp was left at 19C and one deg of increased had already been clicked in.
And, I think that we may not have allowed enough time before clicking up from 20 to 21C.?.

WOBURN WEATHER CENTER - WU Station History
Because it's getting colder now, starting now, I'll only turn the temp back to 20C while sleeping. Same as we did last winter.

That might alleviate these power hiccups.

How often does it happen in the middle of winter?

Not sure, but since the TED is so unreliable, we don't always get an overload alarm.
So, if there is a defrost cycle, followed by a high-power peak at 3AM, it could go unnoticed.

Last season(IIRC), Sanyo #1 had 2 or 3 reset cycles in a row, when the frost got real bad one morning.

Left both units at 20C last night and clicked them up to 21C this morning, without any problem. It is getting colder.

http://ecorenovator.org/forum/attach...1&d=1353424020

The inaccurate TED is showing 11.2 kWh used in 10 hours (since midnight).
But now the sun is coming in, the kWh burn rate will be less.

I missed something. What is the overall goal of the project?

It's discussed back at post #29 & 30


Bare-bones:
Because of the problems experienced with the first unit failing at high power use levels. (In the 3.6kW-plus range).
I do not want my Sanyos to use over 2.4 kW of power. (each).
So, I limit current to 10 amps.(each unit has a limit of 10A).

It seems like Sanyo should make good on this if it's causing reliability problems. You shouldn't have to engineer your own fix for thier issue. If you think the problem is head pressure related have you considered a simple high pressure cutout?