guess who bought a power jack inverter

[BigPines]

OK, you guys are amazing! I just read through the entire thread. I even understood some of it.

I have a couple of PowerJack inverter questions.

1) I have some SunPower SPR-327NE-WHT-D solar panels http://www.stellarsolar.net/downloads/SPR-E20-327NE-WHT.pdf and I'd like to hook them up directly into an inverter. I have read that the high voltage alarm of the PJ LF inverters is set at 50v but the panels put out 64.9 open circuit voltage. I assume there is an easy way to adjust the allowed DC input voltage up by adding some turns to the transformer or by modding the control board to accept a slightly higher voltage. Do I have to do both? Or is there a better DC-DC converter available to bring the panel voltage down to 48v without too many losses?

2) I know the split phase thing may be annoying to many of you but it is unfortunately the reality I live with. The PJ split phase inverters are quite a bit more than their single phase counterparts. How difficult would it be to (and what would be involved in) modifying a standard LF inverter to be split phase? It seems like the split phase transformers probably have an extra 110v winding and it looks like a little daughter card with a small transformer is installed on the controller board to flip the second 110v leg 180 degrees out of phase. I also see what looks like a sensor around one of the AC wires that runs to this daughter card. (see pics) I couldn't find where this card was sold separately. Has anyone done this conversion? If I could get the component list and an idea of where to hook it up, I'm sure I could make it work but I don't think have the skills to figure out what is needed on my own. Any help would be appreciated.

Mike

[oztules]

The pulse width of the waves dictates the output voltage generally. If the transformer is out of range then this is mute... but thats the deal.. so it will operate normally from 46v to about 60 volts easily enough... so thats not the problem... the problem is how the control card sees the over 60 volts.
I have run over 63v without damage, but have no idea how much the input to the control card voltage regulator will go to before damage occurs to the Vreg.

To do that you will have to disable the Vhigh resistor.... I use the 5v6 zenner  in the story.

You will have a pretty feeble inverter with no battery on the front end... not sure why you would do this..


The split phase is simply a center tapped transformer, or two in series on the secondary. The little one and the current transformer is purely for the front screen voltage and current measurements.

So if you have two transformers in it, your good to use them as a split phase output if you want. If it is a single transformer without a split winding, your in need of another transformer to put into parallel - series.. they sell them on ebay.


..............oztules

[BigPines]

Thanks for your response. As you can see, I don't understand all of this too well and have a lot to learn. Thanks for your patience. It is interesting and I'd like to play if I can.

Ah, I don't care about the screen. Good to know that is all the daughter card is for.

The pulse width of the waves dictates the output voltage generally. If the transformer is out of range then this is mute... but thats the deal.. so it will operate normally from 46v to about 60 volts easily enough... so thats not the problem... the problem is how the control card sees the over 60 volts.
I have run over 63v without damage, but have no idea how much the input to the control card voltage regulator will go to before damage occurs to the Vreg.

So I need to fix the transformer windings first to handle 64.9v or is it fine the way it is?

To do that you will have to disable the Vhigh resistor.... I use the 5v6 zenner  in the story.

Yes, I saw that. So I would disable the Vhigh resistor. I'd rather not defeat the high voltage protection altogether but maybe that is the only way. Would I be better off using one of your 8010 controllers? Can the Vhigh be modified on that board somehow?

You will have a pretty feeble inverter with no battery on the front end... not sure why you would do this..

Can you please explain why supplying the inverter with a slightly higher voltage would make it feeble? Or are you saying I need the battery bank to allow the inverter to get the amps it needs on hard starts? I think I am beginning to see.

The split phase is simply a center tapped transformer, or two in series on the secondary. The little one and the current transformer is purely for the front screen voltage and current measurements.

So if you have two transformers in it, your good to use them as a split phase output if you want. If it is a single transformer without a split winding, your in need of another transformer to put into parallel - series.. they sell them on ebay.

Ah, so that is how they do it with a single transformer. They take lead off the center of the secondary winding as neutral. Can multiple center tapped transformers be run in parallel? I assume so as long as you get the polarity correct.

Maybe I should just buy a 6,000w unit which already has the two transformers in it. Do any others have two? I know the 8,000w used to but now it is shipping with only one. I think the 10,000w and 15,000w models have three, correct? I do understand I can buy an 8000w and purchase a matching transformer to add to it.

I don't understand, when you talk about using two regular transformers, the primary winding is wired in parallel and the secondary is wired in series?

BTW, I eventually plan to use the PJ to wake up my 7,000w grid-tie inverter if that makes any difference in this discussion.

[oztules]

Providing your going up in voltage, the transformer will be fine... the pulse widths will just be smaller.

Yes, put two normal transformers in primary parallel, and secondaries in series, and you have a split phase output.  the outputs are floating, so you can put them in series one way, and have two 110v outputs and no 220v or change one around, and have 2 x 110v and 1 x 220v.... so just get the phasing correct.

The 8010 controller won't care what voltage you run it... it is up to you. In it's raw state it does not have upper and lower voltage limits... it just runs with what it has got to work with.

Running without batteries leaves you at the mercy of the sun... and the max power you can draw will be determined by that and the load matching.... that also implies, that your panels can be pulled down rapidly to very low,and your power will be so far from the Vmp, that your losing nearly everything..... suggest not to do it.

I do it with solar pumping units I make, but thats different entirely..... and you do have to make an effort to stay near Vmp, or it won't work usefully.

"I don't understand, when you talk about using two regular transformers, the primary winding is wired in parallel and the secondary is wired in series?"... yes... as above.

"BTW, I eventually plan to use the PJ to wake up my 7,000w grid-tie inverter if that makes any difference in this discussion."..... don't do it without a massive battery bank, and much bigger transformers........


................oztules

[BigPines]

Thanks oztules. I'll get a PJ inverter to play with and then I'm sure I'll have a bunch more questions.

[Offgriddin]

Hi Oztules and others, I have been unable to get our PJ 8000w to work as a battery charger. Does anyone have experience with this? The manual is woeful.

So far my battery charge procedure has been:
1. Power down PJ
2. Connect generator output (6.8kva genset) to PJ 240v input using wired terminals
3. Start up genset and turn genset power output on
4. Turn on PJ switch
5. PJ boots up no charging happens, just outputs as inverter.

Aditional question, manual shows PJ being used as a UPS type arrangement, as in can run 240v input in the PJ and when that fails the PJ will automatically pull from the batts, while maintaining a continuous 240v output. Is this correct? A lot of the warnings about the PJ tell you to power down before changing from inverter to charger function, but is it ok to go from charger to inverter function with powering down?

A walkthrough from anyone who has used it as a charger would be most helpful.

Our inverter is 2015 PJ 8000w with two torroids.

Thanks in advance!

[oztules]

I don't do it any more, but suspect I have blown more boards up than most trying combinations.
I have never had a problem with destructive behavior when the pj syncs to the genny... so as far as my playing has been concerned, you can run the genny, and hook it up while running as an inverter, as it gently syncs to the genny frequency. It has to do this even if you turn the pj off first, the same events need to be enacted... it is the switch off that stinks.... make sure you turn of PJ before disconnecting source/genny.

Not syncing is usually only due to voltage of genny out of range, or frequency out of range.

If they are in range it will/should pull in.... so check your genny voltage and frequency... they may have one and not the other.

If AVR, the voltage may well be right, but rpm wrong, and the same for cap excited units... for different reasons, but same effect.

I have a W7 ( same cards essentially) that I tried for many times to get the actual grid to connect, and it would not.... but it connected to a genny one day I was fixing for someone else.... the grid was not high enough... so it thought anyway.... tested theory by using an auto transformer winding, and bingo, worked every time.

The W7 did not need disconnecting before switch off... and this caused me to experiment with the PJ too much. ( W7 has EI transformer)

It was not that the grid was so bad, but the PJ/W7voltage measuring staircase was iffy at best.


............oztules

[Offgriddin]

Thanks for the reply Oztules. Interesting...as always. Wil check some specifications and try a couple of startup procedures.

[UFA.Solar]

Hello to all here in the Forumhttp://www.anotherpower.com/board/Smileys/default/kiss.gif
I am a new bee and have to say it's great what I  read over the whole thread
according the LF inverters and the W7 modifications.
Congratulations to You sharing all this  use full  information in this case. 
I came on that forum across the W7 activities in an parallel forum.
I found some time ago in an German forum about Photovoltaic systems an hint
to You guy's and all Your activities.
Now to my  little project :http://www.anotherpower.com/board/Smileys/default/huh.gif
On the roof of my house sitting 36 Panels with prox. 9.8 Kwh/peek.
Via an three phase grid tied SMA 9000 converter the power is used  partial
for own consumption and the rest runs over an two directional recording
counter into the main grid. The earnings due to German regulation are 0.15€
per Kwh . In the rules is defined , the own consumption has to bee 70% of
solar earnings. So I put an logger into the line . The logger counts the
output from the grid inverter plus the readings from the two way counter.
This system is now running since  Feb. 2012 with no issue.
I found out that my consumption is very close to the 30% depending on
the weather condition in the year.
In spring & winter no problem, but in summer & autumn I am out of spec.
So I put up an cheap inverter for 24V ( brand name Westech Combi S ) which is
as I found out later similar the W7. It draw in idle 2.5 to 3 A in my
case. So not good to run cont. The good thing is, the included loader
which gives at 24V aprox. 30 to 35A current in position 100% load.
There is a 3 digi sw for current reduction to 25, 50 and 75% , which is
very use full in my case. Please see later why !
The other goody it CCCV+time load characteristic and able to connect LiFePo4 batteries.
I use a 270Ah 8s3p 24V System with top balancing each cell and have
a home brew battery watch system with upper and lower limits for the overall
voltage ( currently 30.2V high end 24.2V low end) controlled by a small
AVR system including net availability in the local network. It also
saves data on SHDC and addition every minute in a data bank
on one 24h's running separate raspberry-pi II with a SSD drive joined.
So I have access to the status over long periods and get some idea what's going on.
It is also possible, when I am on the road to get on the system even through the
internet via tunnel application.
The inverter is only used on one phase which is switched  from a  program
running in the background of the UNIX driven solar-logger.
The signals are applied by relays which are driven on the lpt-port from
the logger.
Simply spoken, if their is power going from the solar grid inverter into
the mains grid and it is more than 500w , the program start the
loader part in the W7 inverter at 50% current ; is the power > 1000W -> 75%
and  more then 1500W go to 100% load current. It also counts the charge time
accordingly. That gives me back how long the battery can be discharged
safely also on week sunny days at winter season
The use time is during the night. I switch the inverter 2 minutes
on ( very low time idle ) before I disconnect the phase "T" from the grid
and change this line to the inverter output. The length of "on-time"
is calculated from the load-time * a  total efficiency  factor found over
a while using the system. It is possible to change most parameters of
the system on the fly via the network.
Typical consumption is 1,4 to 1,8 Kwh over the whole time period .generated by
the inverter. The consumers are deep freezer, freezers, circulating pumps
from the heating system, all the server's and TV-sets and lights at the
staircase and basement . This system is now alive since May 2014 and saves
the 70% problematic. At the time its running aprox. 35-37% .
Now my questions:
I bought a LF5000 24V a month ago . The unit was real real cheap < 300€ !!
made some tests and found idle current is 0.85A , what seems to be ok.
the round choke with 3 or 3,5 turns is in.
But the loader work's different the W7 type . only high load current
continuous going down over the time to an float stage. The PCB (controller)
looks very similar to the W7 with a few changes on fan control  NTC's including
a switch for continuous run at the rear.
Fancy display with current and voltage sensors are also attached.
A: Is their somebody having a schematic about the upper PCB for the W7 and the LF series ??
please let me know .
1: Is the program in the case W7 different to the LF series inverters ??
2: Do You think this new LF5000 will do the job in my case ?? (typical load 300 to 700W)
3: Is their any one tracing a similar concept as I have in place. will try to get
a sketch up into this thread.
Any solution is appreciated
UFA.Solar
Sorry for German at the schematics produced some time ago.
Over all diagram plus details

[whisperingsage]

I bought the 24 VAC 8000 watt power jack based on Oztules' post. I am off grid. I ave been using square wave inverters for years due to cost, but found out the square wave was burning out my refrigerators, got a bunch of dead ones I use for storage now. So when I saw your review on te Power Jack, I got one, as we had  been struggling with a Harbor Freight square wave 6000 watt inverter that never gave us the power it said. Haven't been able to run my deep well pump for years, not since the AIMS, which was square wave, but at least ran the well pump for 4 hours a day if we wanted. (We store water in a 350 gallon tank and must pump for garden and automatic gravity waterers for livestock. )But I couldn't get my husband to help me fix the huge hole in te pumphoouse roof, and it rained, burning out the AIMS and throwing us back into primitive living for YEARS. Wasting money on gasoline for generator and noise.

Anyway, the Power jack I got only has one big torroid in it, and I think they learned from the windings on the magnet as a choke when I went to put mine in, (made from Oztules' instructions, ) they had placed a small one in there, just where he said to, only smaller than his (and mine) and on a small torroid magnet, not an E like I had made (like Oztules had) so I thought, well, nice! The wires they use are about 10 AWG while mine was 4 AWG (the one I didn't use) And it worked nice for three days. We have 10 deep cycle 12 volt batteries, wired in series to 24 volt, I have 1400 watts solar going in and 2 Outback controllers, which we adjusted to 24 volt system, There are 4 battery terminals which confused me for awhile until it looked like i only needed to hook up two. The one thing I don't like about it is the LED screen is easy to read when I first turn it on, but the backlight goes out in a minute and ten I can't see a thing as to what watts and amps  and volts are going on!!. While it is on, it does appear that watts stay about 45 at idle. So that's good. I was able to run the pump AND cook on the griddle at the same time about noon, good sun. So I was pleased with that. I tested it on the small fridge (235 startup- and 104 watts working power according to the killowatt meter- and I plan to glue 2" insulation all around the outside except the area were the condenser is, and this keeps it cooler longer. Uses less power. ) and the fridge worked fine.

But it acted weird today. It pulls my water pump fine- 1400 start and 1200 continuous, and it will run it for hours as long as the sun is out and sky is clear. But when i turn the pump off, the small load from the laptops and LED lights are no longer seeming to keep it on, it does the blinking thing. Irritating, as I have a lot of low power stuff I need on. It was fine for a few nights there. Don't know what I did wrong. I fiddled with the inputs and the plugs and tried to see if something was unplugged, but no, then I plugged in te fridge and that got it working again- it seems it liked higher power even a few hundred watts, rather than lower power. I thought it had fixed itself until I unplugged the fridge to move it inside and the inverter went all screwy again. Back on generator at night until I get this troubleshot. Any ideas?

[rossw]

I don't have one of these, but it sounds to me like you have it in "Automatic" mode - it's "searching for a load". With a few very small loads, it's shutting down.

If you can find it's "mode" switch and select "always on" rather than automatic mode, it'll probably fix your problem.

[oztules]

I agree with Ross. try the other position on the big rocker switch. It needs at least 50w on the search mode I think to make it run..... would normally be used to run the fridge or fridges only, and turn off the rest of the time.... you can save idle power that way, and the way they were originally that coulds save you 5kwh/day. power that way....

I have never bothered to use it like that for more than a few minutes as a curiosity.

The little torroid makes a huge difference, not quite as much as the ecore does, but good all the same.

Edit: A 100r resistor across the back of the LCD screen will keep it on all the time... I will take pics of the mod when I can get one out. 

...............oztules

[oztules]

UFA.
The W7 is slightly different to the PJ control card... they are sort of interchangeable, but small differences. The later model W7 had variable current charging, the PJ does not. You can only select a range of different battery technologies.

It will drive your loads ok, but the charging is very iffy, and you may blow it up if you don't switch off the PJ before turning off the grid going to it. The W7 runs this part fine as it has different transformer types.

The W7 transformer has lots of leakage, so handles the frequency change from charge mains freq to 50 hz on change back to invert.. the torroids tend to blow things up as there is little to no leakage there to act as a buffer.

Circuits are not available as far as I know for the control card, except for the driver stage.

With no current control, and patchy turn off characteristics, it may not be what you want... if you have a separate charger it will do fine.


.............oztules

[UFA.Solar]

Hi Oz
many thanks about Your verry quick answer !!!http://www.anotherpower.com/board/Smileys/default/wink.gif
The information from You is helpfull .
Yes the W7 is a different type of transformer big M-core , which is using up a lot of current idle.
In the data sheet I got with the unit theye write :
Charging curve ( 4 stage constant current ) Battery types ;
4-step digital controlled progressive charging see the picts attached.
When i look to the different boards there is still on the new LF-PCB the terminals for the
current dig sw ; only no cable . so I put up an variable resitor to check out if there is some
change in output current --- nothing changed -- .
My W7 is in productive phase and I could not look to details on the PCB what is the real
difference.  Do You have some idea . I put a detail pic from that area attached.
UFA.Solar
details digi-sw PICT1394

[off the wall]

UFA Solar - what an amazing system you've put together. Thanks so much for taking the time to give such detail and I'm going to be studying it carefully.

For those not wanting to go into such complexity, however, system modelling charge with a 5W or 10W solar panel feeding a 500F capacitor and modelling discharge with a current transformer modelling load driving a LED loading the capacitor with an LDR can give a very rough and ready but useful indication of charge/discharge balance.

Best wishes

OTW