My install and review of the AIMS Prius 2kW Pure Sine Wave Inverter for Backup Power Generator

Oh, I think you have done an excellent job of researching this installation and have definitely come up with a safe installation. I really appreciate your write-up and discussions with Bob over the proper fuse for the application.

I look forward to installing mine and testing it. I will let you know how it goes.

On your power indicator that you have sent back for repair, is this the green indicating light?

 

I connected my inverter today to check it out and make sure everything was working before the fuse holders and fuses came in. The green LED power indicator worked correctly on mine. Sorry to hear about yours.

The 2010 Liftback battery connections are slightly different than the PIP. The + white wire is connected with a right angle connector. This made it slightly harder to make the inverter connections. Plus the fact that the direct connection wires that come with the inverter have large ring terminals on both ends. I will use smaller ring terminals on my harness that I make up.

I connected a refrigerator I have in the garage and ran it for a while. I lowered the thermostat setting to get the compressor to come on which it did. The inverter output stayed fairly constant at about 118 VAC through the full range of the battery from 220-200 VDC. I let the ICE cycle on a couple of times. It cycled on at 2 bars and off at 4 bars much like when you drive in stop and go traffic and use all EV.

I am looking forward to getting my supplies so I can make up my harness similar to what you did. Thank you so much for all of your information and research.

 

AHetaFan,

First let me apologize if I am hijacking your thread. You have so much good information and research on this inverter installation that it just makes sense to me to keep this all in one thread for others to review.

I followed your installation and used two inline fuses with Ferraz Shawmut A60Q15-2 semiconductor fuses in the wiring harness. Since I have the 2010 Liftback I ran my harness to the cargo area and will store it in the 12V battery compartment. I wanted to keep the fuses as close to the HV battery DC output connections so ended up installing them in the Main Relay compartment. See attached pictures.

When I powered up the inverter and began running a small table fan I got a surprise that I was not expecting. I heard the car beep and looked up to find a Check Hybrid System alarm message I was disappointed that the SGII would not display a DTC message but instead said No Pending Codes. I could clear the alarm but could not get it to display the DTC.

After an hour or so of troubleshooting including removing and installing the HV battery disconnect plug several times, reviewing the harness wiring and connections, battery quick connect plugs, etc. I determined that I received the message only when the inverter was connected, even when the output was turned off. If the inverter was disconnected the alarm message did not return.

I have sent an email to Jack Chen with the inverter supplier. I will keep everyone posted of the outcome.

 

Any chance the orange battery disconnect wasn't entirely seated correctly? It has two connectors, one large one for the battery and one small one which tells the computer if the disconnect is in place. That small one can be hard to reseat properly and will throw a warning. Doesn't sound like your issue, but checking just in case.

 

First thing I went to and opened and re-closed several times. With the small interlock connector being on the back side of the plug it really is hard to tell if it is seated properly or not. I even removed the interlock wiring connector from the stationary plug frame to check the wiring and the male and female pins to make sure they were OK.

It was after I convinced myself that the disconnect plug interlock was not my problem that I disconnected the inverter and realized that I only received the alarm message when the inverter was connected.

Thanks

 

I wonder if the ground-fault detection circuit is sensing the inverter?

From memory (I'll check my sources), the earlier ground fault detector uses a 5 MHz signal to detect ground faults. Do you have an oscope to see if there are any high frequency signals on the DC lines?

If you have a high-voltage cap, see if putting it on the connector replicates the problem. It may be a simple problem of a choke to block the ground fault signal. But we need to know the frequency to 'do the sums.'

The other hypothesis is 'inrush'. The 12V inverters have caps that have to be charged when first attached. Although not perfect, a choke can temper the inrush. Can you measure the capacitance of the inverter?

Was the inverter connected to chassis ground?

Just a thought,
Bob Wilson

 

Thanks for all the info and comments -- this is very interesting. I have one question. It's very nearly impossible to determine the state of charge of NiMH batteries (except when fully charged or discharged) by looking at the voltage. So I've assumed that the Prius must keep track of the charge going into and out of the battery in order to know when to charge it, and the approximate state of charge. Does it? If so, is the inverter hooked up at a place where the car can detect the charge going out and account for it? I'm thinking that if the inverter is hooked directly to the battery, the car might lose track of the state of charge, and not charge the battery correctly when the external load is hooked up. Does this happen?

 

We don't know for sure but all data suggests they are not 'coulomb counting.' Otherwise, the early plug-in and Richard's UPS Prius would not have worked.

One of the early plug-in problems were using relays that spiked the traction battery. Relay close would spike the traction battery and trip various various fail-safe codes. Eventually folks realized the right answer was to use a DC-to-DC inverter and ramp up the charge. They had no problem adding charge from the 'plug-in' batteries to the NHW20s.

However, I've not read how the ground-fault detector was defeated. A high-frequency filter would work and a pair of chokes could do it. Just I don't know.

Bob Wilson

 

That is what I was thinking too. I plan to measure the + and - terminals on the inverter to the inverter case and to chassis ground and see what the resistance values are. Jack Chen, with the supplier, says they should be > 1 Megohm.

I have an oscope at work that I can bring home and check this.

The interesting thing about the alarm message is that it did not come on until exactly 1 minute and 15 seconds after the car was in Ready mode or after resetting the alarm with the SGII.

No, but the inverter was laying on top of the HV battery metal case so had metal-to-metal contact. Last weekend I had the inverter hooked up temporarily and laying on top of the carpeted cargo deck floor and did not get the alarm message.

Thanks for the input. I will keep you posted of what I find.

Dwight

 

Last night, I re-read the section in ground fault detection. The only thing Vol. 1, Diagnostics, had to say was use a 500V meger and see 10 Mohm between the battery buss and ground. I could not find a reference to the ground fault signal frequency in the diagnostic manual. The 5 Mhz comes from "New Car Features" referenced in a "Prius Technical Stuff" post.

That it took a minute or so before the fault showed up is consistent with ground fault detection. It would make sense to let the car 'warm up' a little to let any fog or condensation evaporate before the ground fault test.

Let me suggest repeating the test with the inverter insulated from the car body. Then measure the voltage and signals between the car ground and inverter chassis. Use a variable resistance to find the lowest value that triggers the fault. That value x10 would be half of the inductor reactance needed at 5 Mhz on each DC leg . . . or just make sure the inverter chassis is always insulated yet still cool. . . . Heat is the enemy.

Bob Wilson

 

No, I did not take the time to run the refrigerator again. The fan would run even after getting the warning alarm message from the car.

I had the inverter isolated from ground the first time I tested it when it was sitting on the carpeted cargo deck. This time it was sitting on top of the HV battery metal housing so it was not isolated.

See my post below about the explanation for the alarm messages.

 

I wish I had taken the time to test the resistance-to-ground of the positive and negative terminals of the inverter unit yesterday but I ran out of time. I really wasn't expecting anything this simple either so I contacted Jack Chen first.

I checked the inverter unit after getting back from church this morning and the negative terminal to the case ground is reading a dead short. The positive terminal is reading above the 200 Mohm scale of my meter. Jack says they both should be > 1 Megohm.

No wonder the car is triggering the Check Hybrid System warning message when it is seeing a dead short on the negative side of the DC.

When I tested the unit the first time last weekend I had the unit isolated from chassis ground by sitting it on the carpeted cargo deck. This time I had it resting on the HV battery metal case so it was making metal-to-metal contact to a grounded chassis frame.

I have sent an email to Jack Chen to see if he will let me open the case to have a quick look inside without voiding the warranty.

 

This is why fusing both DC lines makes a lot of sense. Fuses on both feeds protects the wire. But I have to admit, finding the inverter ground tied to one side makes me reconsider how I would mount it.

Our cars are 'in the rain' and I don't cotton to rain-wet tires providing a path to earth ground. I really do not want the traction battery, B- having a ground path. It only takes unit milliamps to stop a heart.

I do not mind if the AC neutral is ground but I have to think about the DC negitive. It may not be a problem but I need to think about it. For example, would a rain induced, ground fault lead to the car detecting and reporting an error?

New information, the analysis needs a fresh look. It could be a non-problem. I'd like Richard's thoughts too as well as the "Prius Technical Stuff" brain-trust.

Bob Wilson

 

Yes, I was really surprised too. According to Jack Chen this should not be as both terminals should be isolated to ground. I am hoping that he will let me open it and I would love to find an inadvertent ground wire to the negative terminal that can easily be removed.

I was going to suggest having them check yours before returning. Since you will get it back tomorrow I would definitely check it.