PIP inverter repairs and hardware modifications

[coulomb]

[offgridQLD]

I think I need to make you guys some custom spanners (basically heat up a cheap few $2 store spanners and bend it into a funky shape to make easy work of jobs like that.

What size is the bolt head holding the fets in?

Kurt

[weber]

I was, quite rightly, berated by Coulomb for not having Z (or offset) screwdrivers, which of course he has in his workshop.

The heads on those M3 screws are 5.5 mm hex and #2 phillips. The white capacitor definitely has to come out. It has maybe 1 mm of clearance. Even if you could turn the screws there's no room for them to come out. But the black capacitor has about 20 mm clearance from the screws which are under about a 10 mm overhang of heatsink, so a suitable offset screwdriver or open-end spanner could do the job there.

I also don't have open-enders below 8 mm (only sockets). Coulomb also pointed out that I need a miniature file set. I feel a tool shopping expedition coming on. Is there anything else I've forgotten, Coulomb?

[Scott]

Someone on mybroadband.co.za was nice enough to take a photo of the back of his 2015 comms board.



The IC is labelled as a SC510070JDME, but after some poking around it seems to be a MC68HC908JB8. I thought it'd be a generic usb-serial chip but it's a small microcontroller.

On the plus side it looks like the pinouts on the RJ45 are the same, with +12V on pin 4. RS232 works as per normal apparently.

[coulomb]

Someone on mybroadband.co.za was nice enough to take a photo of the back of his 2015 comms board.

Neat. I might add that to your post with the 2012/13/14 comms board photos. [ Edit: oops, I see you've already done it. Thanks! ]

So this board has a D9 for serial, a USB socket, and the dry contact terminals, left to right in the photo.

Good to see that RS232 seems to be unaffected.

[Scott]

So this board has a D9 for serial, a USB socket, and the dry contact terminals, left to right in the photo.

It's the same RJ45 connector for serial, and a USB type B connector.
The inverter ships with the same RJ45 to DB9 connector, as well as a USB A-B cable.

[coulomb]

The inverter ships with the same RJ45 to DB9 connector, as well as a USB A-B cable.

So what's the USB port for? Can you run the WatchPower software through USB without a USB dongle? Does it have to be a recent version of WatchPower if so?

[Scott]

The inverter ships with the same RJ45 to DB9 connector, as well as a USB A-B cable.

So what's the USB port for? Can you run the WatchPower software through USB without a USB dongle? Does it have to be a recent version of WatchPower if so?

I don't know, sorry. The guy I was speaking to wasn't able to get the drivers installed on mac, and didn't bother too much with it because serial was easier.

[coulomb]

So dockarl, we think we're getting close to repairing the two boards.

All the parts have been installed, and the 2013 (original) board is back up and working. The 2014 seemed to need some more parts, they've been ordered and installed.

We tested the 2014 board today. It powered up fine from the mains, and then from the battery (turning on the AC output switch). But as soon
as it started to run the inverter, things did not go well:



You can see in the circled area that one of the MOSFETs blew off two of its leads. Sigh. Weber and I were both quite confident that everything would work. There must have been more damage to the MOSFET drivers (from the original fault), that was masked by the power supply not working.

This board seems to be in a worse state than when Weber received it initially. I'd say it's not worth the effort to get it working any more.

At least the original board seems to be working fine. We hadn't tested it with the SCC since the repair, so today we hooked it up. It charged, but unfortunately, it was by then quite cloudy, and it's now steady rain. So we never got to see much solar charging; nothing over 3 A. Hopefully Weber can test that over the next day or two.

Let's switch to email to complete this.

[ Edit: reworded a little. ]

[rinaldoparaipan]

Hello everyone.
It seems that the new type of 4048MS(which is build in a new case, without the black heatsink at the upper side)has problems with the 3300uF capacitors.
I had problems with my inverter, a friend of mine has the same fault: with only 500W load the inverter collapsed after 5min.The capacitors exploded and the device stopped functioning.

[paulvk]

Is it the only solar charger in the system?

[rinaldoparaipan]

Yes, is the only one.

[paulvk]

What type of batteries are you using.

[offgridQLD]

So no big black heat sink up the top to cool the MPPT solar charge controller.

How is the solar charge controller cooled then?

Edit: Or is it only a PWM solar charge controller so not as much heat to deal with Like the MPPT models.

Kurt

[paulvk]

My unit is not getting hot at all even with 60amps of solar so the heat sink has to be overkill.

[offgridQLD]

Are you sure it's MPPT and not PWM. If it's MPPT there has to be some kind of aluminum block at the top perhaps just with no fins?

There would have to be some way to dissipate heat from the Mosfets.

Pics would be great.

Kurt

[paulvk]

My ones are the units with the heat sinks on the top but only get warm even at 60amps they are MPPT units.

[rinaldoparaipan]

The unit is MPPT.When I opened the box i thought that is a wrong delivery, but the label is MS 4048.MPP solar confirmed that they changed the case after July 2015.
The problem is the MPPT board, the compacted construction and the capacitors very close to the heatsink.

[fllniks]

It will be best if you can click a pic of the unit and side sticker on the right side of the inverter. Post those pictures here.

[rinaldoparaipan]

Here are the pictures

[paulvk]

It looks like a different MPPT unit I wonder if the older ones are capable of much higher currents.
All the rest looks the same.
You sill have not said what type of battery, is it lead acid or something else.
The two caps that have gone would be across the battery and at 63volts I think there is a need to keep the system at 58v max I have set my ones to 56v when the warranties are up I will be changing them to 100v.

[offgridQLD]

Interesting,
            Thanks for the pics. I wonder if the decision to use a thin strip of aluminum plate at the MPPT charge controller as the heat sink was a cost cutting decision (BIG heat sinks are expensive)

As others have mentioned, what size and kind of battery are you using. How much solar do you have connected and at what voltage?

Also how old is the unit (how long has it been in service)?

Kurt

[paulvk]

I have a question for those that have taken their units apart and looked closely at the circuit and components as I still have warranty I have not completely pulled my ones appart.
Other than the 63V capacitors is there any reason it could not go to 62volts ?
I know its firmware will not but what about the components on the DC side.

[coulomb]

Other than the 63V capacitors is there any reason it could not go to 62volts ?
I know its firmware will not but what about the components on the DC side.

There is the voltage rating of the DC-side MOSFETs. According to the "repairing service guide", these are supplied as 11-320110-00G or IR/IRFB3307ZPBF which are 120 A and 75 V rated. 75 V does not leave a whole lot of headroom for inductive spikes.

According to this post by Solamahn, later models are supplied with higher current MOSFETs, but still 75 V or 80 V rated.

If you routinely need to run your DC-side at 62 volts, I'd suggest replacing those MOSFETs with at least 100 V rated parts. Weber posted about upgrading MOSFETs as well as capacitors here.

[offgridQLD]

Hi

I have the 2.4 KW model running a VFD controlled pool pump and a Inverter Aircon (12000btu).

No issues at all.

That's good to know thanks.

Has anyone had the need to remote activate or more simply switch on/off there PIP from a distance. I'm not talking away from the location over the internet. Just say 30 meters away.

I intend to use my pip in the future just to run my hard wired EVSE (electric car charging) Though I don't want to have the unit turned on 24/7 at 50w idle consumption ( 1.2kwhr pr day) As some days it wont be used and days when it is it would only be for a few hrs.

I looked at the small rocker switch at the bottom of the PIP and wondered if there would be any harm in extending that to the garage and placing a switch on the wall near the car charger. It has to be simple so anyone can use it.

It's just not practical to walk to the other end of the house. Unlock the power room door and switch on the inverter. But snakes, rain, cold, hot all make it less than practical at night.

I Have remote coms to the room (as in the room is part of my home network via wifi and LAN connection ) though want to keep this simple so was look at the switch.

Edit: Yes I did see the option remote unit sold for the PIP inverters but it lists it as

"Maximum operating distance of the remote module is 15m connected via RJ45 network cable. "

Though perhaps some kind of powered hub could be used to extend that distance?

Kurt