Grounding kit for the Prius - would it make sense?

Over time, ground points do become corroded - especially where there is a lot of road salt - and cables become corroded as well. On many older vehicles I've had good luck every 2 years taking the ground connectors apart, smearing silicone dielectric grease on the ground surface and bolt, and reassembling them.

Note: the battery ground cable MUST be disconnected while doing this or you could risk a spark or damaged components.

In particular, my 1992 Honda Prelude SR started to run poorly in 1996. The dealer was clueless. The shop manual did mention, rather far down the list, to check for grounds.

It seemed intuitive to me to use my DMM set on VDC, put one lead on the battery negative post, and start touching the other lead to the body and the motor while the motor was running.

With the motor running, I had almost 6 vdc when I touched the motor.

I went to Canadian Tire and got a small length of 6 ga battery wire with lugs preattached to both ends. I hooked up one end to an accessory bracket on the motor and the other end to the negative post. All the problems went away.

Oddly enough, when I went over all the body and engine grounds, I couldn't find the source of the trouble. I took apart and cleaned all the contacts, check the ground cables for continuity, yet there was still obviously a ground problem somewhere. So I had to leave that jury-rigged ground in place.

With the Prius, there already appears to be a good grounding method integral to the HSD system. That's necessary when you consider the high voltages and currents involved, there is a lot of power there.

One thing I would caution that if you happen to hook up a ground to a point the engineers didn't anticipate, the factory Ground Fault Monitor might activate and kill the high voltage system.

That's a safety feature as the Ground Fault monitor assumes a voltage potential between the isolated high voltage system and the body indicates voltage leakage. For safety, it then opens the relays.

Otherwise, for a new vehicle I question if an aftermarket grounding system is necessary. Usually the factory method on a new car is satisfactory.

 

The DC/DC converter is integrated with the inverter, which is bolted to the PSD assembly where the ICE is also bolted. That would eleminate any problems with ground problems with ICE. I think you would get more resistance from the 12V connections than the 12V connections.

However, there is the matter of body ground, where the ECUs get their ground. This is where ground loss could raise havoc, and might benefit a redundant ground strap, though its benefit might not be apparent for a few years, as original connnections start to deteriorate.
The battery has a 150A fusable link, the DC/DC converter has a 100A fusable link. Either can supply current to the 12V electrical system, depending what the needs are, and the power mode. When not in READY mode, I typlically see 11-12V, in READY mode, it is closer to 14V.

So, the ground straps would benefit accessories more than it would benefit the ICE.

 

<div class='quotetop'>QUOTE(DanMan32\";p=\"87818)</div>

Dan:

Sorry what I meant was that when I touched one lead from my DMM to the negative post on the battery, and the other lead to *anywhere* on the motor of my Prelude, the meter registered around 6 vdc.

In theory, the meter should have registered 0 vdc, since I was measuring from the negative post, not the positive post. The implication is a bad ground from the motor/body to the negative post of the battery.

When I got that chunk of starter cable, hooked one end up to the motor accessory bracket, and the other end up to the negative post, the situation changed. Touching one lead to the negative post and the other lead to anywhere on the motor registered 0 vdc as I would have expected.

The car ran much better, headlights brighter, etc. I never did figure out where I had a bad ground, so I left it like that.

A regular automotive battery at float or rest should register around 11-12 vdc. In READY the inverter/converter is supplying charge to the auxiliary 12 vdc battery so the voltage should be at least 13.8 vdc.

I do foresee the possibility of the "grounding kit" confusing the Ground Fault safety monitor and causing the HV relay to open. For that reason alone I wouldn't try such a kit on the Prius.

Jay

 

To rookie: After reading Jay’s posts, if you are still up for it, you’re more than welcome to take my free trial offer whenever you & more importantly, your Prius are ready. I recommend that you don’t install the grounding kit until well after the first oil change at 5000 miles.

 

Sorry Jaman if you thought I was criticizing you. Your means of diagnostics is quite sound, and I commend you for figuring that out. V=IR where in your case with your Honda, I is the current flowing through the wire, and R is the resistance the wire presented, thus giving the voltage reading you found.

I just don't see faulty ground being or even becoming a problem in the Prius, not that it coudn't happen.

Using a voltmeter would be a more conclusive way to determine effectiveness of ground straps: voltage from engine block to car body, voltage from car body to aux battery negative terminal, and from battery negative terminal to engine block. Test under highest current draw. mV can be from ambient elecctromagnetic radiation from the car and outside sources, as well as drift in the meter itself. DMMs have very high impedance so they can easily pick up stray currents, even from your own body if you touch both terminals.

 

<div class='quotetop'>QUOTE(DanMan32\";p=\"87886)</div>

Dan:

I agree, the Toyota engineers have designed for a high voltage and high current system.

Out of curiosity I took my high-impedance Fluke DMM down to the parking garage and powered on the car. Under the hood, I poked around by touching one lead anywhere on the motor, the other lead at different electrical and body ground points.

At the most, I recorded around 50 mV DC. That certainly is within a tolerable and expected value. I just don't see the need for a "grounding kit."

Again, I would worry that inadvertent application of a "grounding kit" would confuse the Ground Fault safety monitor. If there actually was a poor ground, and you didn't properly diagnose and repair it, just using the "grounding kit" will more than likely trigger the Ground Fault system to open the HV relays.

Jay

 

50mV, OK but how much current was flowing when you measured this voltage?

HV is not relevant here, all leads are isolated from ground.

 

If I recall, my Fluke indicated a *very* low current, something like 0.1 mA. The meter is at home and I stored the values, will double-check when I get home.

Actually, HV is relevant. I would expect a certain amount of inductive pickup and transfer. Nothing huge and just taking a quick peek underneath the car a couple of times, Toyota appears to have taken this into account when they ran the HV cables.

 

Fluke: you get what you pay for. In this case a memory to record tests.

The Prius registered 0.1 mA and 48 mVDC. IOW the readings are insignificant.

I use the Fluke more for industrial process control troubleshooting. It has extremely high impedance, is Intrinsically Safe, can power 4-20 mA and HART process loops, and can display simple scope trace.

You'd be surprised how often a problematic process loop is caused by a floating ground or bad corrosion of terminals..

 

You weren't driving enough current to make a determination. Try turning on the headlights and the windshield wipers, then measure the voltage across your ground straps. But with V=IR or R=V/I or 48mV/.1ma that's 480 ohms which is quite a bit. You won't get the needed 80-100A out of that resistance. A dead short circuit through it (12V) would only yield 25mA.

 

By the way, I was not referring to the current through your meter, but rather the current through the ground wires you were measuring across. If .1mA was the current through your meter while you were taking a voltage reading, then you have a very low impedance meter. Most digital meters have impedances on the order of tens of megaohms. Analog meters are usually 50KOhms average.

 

I recall a few years back when I frequented the Echo fans website, that some of the kids there put grounding kits on and reported better performance and MPG. The Prius engine is basically an Atkinson cycle version of the 1NZ-FE in the Echo so it just might help. Only things that might keep someone from doing this is the tight spacing in the engine compartment and all the other electrical stuff you have to work around.

 

It isn't a matter of the engine, but how well the engine is wired. Siince the 12V is primarily coming from the DC/DC converter that shares the same grounding block, and the DC/DC converter would provide clean power, I doubt a ground strap would help any. In the Echo, the clean power would be coming from the battery who's ground terminal isn't even bolted to the frame, let alone the engine block.

 

Dan:

Ok, what sort of values should I "expect?"

I have a Fluke 189 recording DMM with 1 microvolt resolution, a Fluke 744 with HART, and a Fluke 430 Power Quality Analyzer. I used the 189 to check for voltage potential between the motor and the unitbody.

I was looking for any *potential* difference, not actual current flow or voltages. This method of checking has wide use in industrial and commercial applications.

To check for actual current flow, I would had to have disconnected the negative lead at the NiMH battery pack and put an ammeter in series. That could create a floating ground and easily destroy the IGBT in the inverter.

There are several ways to check for ground continuity and electrical leakage. All have inherent positives and drawbacks.

The most popular method is to use a Megger but with the widespread use of solid-state components, that is now discouraged. A megger will easily destroy solid-state components unless they're removed from the circuit. The megger is still used to check rotor/stator and wire/conduit insulation continuity.

Another method is to check, in realtime, the ground continuity at one or preferably several points around a circuit. If any test point develops unusually high resistance, possibly indicating a broken conduit ground, the possibility of neutral flow or phase imbalance is implied.

Another method is to monitor for voltage *potential* between the "true" ground and various ground points. If voltage *potential* is discovered one has to assume poor ground continuity and voltage leakage. This has widespread industrial use with Ground Fault monitoring.

There are portable test devices that will even determine if your home ground is still functional: more often than not you'll be quite surprised with the results of your supposedly "grounded" home. Eg: some folks will flip the breaker to OFF before working on a circuit, and they'll use a DMM to ensure "zero" voltage.

They're often surprised to find up to 20-40 VAC between ground and either Neutral (Unbonded neutral service) or Hot. Guess what? Your "ground" either is no longer a "ground" or you're serviced by a 3-phase transformer (Common in a multilevel building like a highrise condo) and the phases are poorly balanced. Either way, even with the breaker OFF, you could get a tingle or worse.

I do a lot of industrial work and am very familiar with Reliance motor drives. I also have access to the Rockwell Industrial database. Please reference the following installation and procedural manuals:

http://www.reliance.com/pdf/drives/instruc...uals/D23360.pdf

http://www.ab.com/industrialcontrols/produ...1409-wiring.pdf

http://domino.automation.rockwell.com/appl...5256B79005AA62F

http://domino.automation.rockwell.com/appl...5256AFB006F8A6B

 

Dan:

I just got back from my local Toyota dealer and they had an Echo Hatchback in the showroom, so I didn't have to get soaked in the rain.

The ground from the battery clearly goes to the accessory bracket, then a smaller strap is evident to the unitbody.

Jay

 

the high voltage has NOTHING to do with ground. Both leads of the HV battery are sent to the inverter via two separate cables: one for +, one for -.

We are talking about checking for 12V loss due to ground loss here.
In order to find the resistance between two points by the method of taking voltage readings at those two points, you have to know the current flowing through those two points.
If you have an inductance current meter, then you can find out what current is going through the ground strap that you want to find the resistance of. Then measure the voltage at both ends of the strap, or better outside of the connections of the strap so you are also including the junction for the total resistance.
Then R=V/I. If you find the current flow is 30A and the voltage is .01V, then the resistance is .0003ohms. Or better yet, for the current load, the load is losing only .01V from the source, or at least the source for the strap.

Besides losses through ground, you can also lose 12V power from the + side.

 

You mean the Prius has a "floating" ground for the HV side and a conventional bonded-to-body ground for the 12 VDC side? Hmmm. I wasn't aware the inverter/converter was isolated in that way. Could you please post the appropriate wiring diagram?

All I have to go on is the 2004 Emergency Response Guide, P 16/26, which is somewhat unclear about the negative side. It appears to indicate that the Ground Fault system monitors the positive side and compares to the reference negative side and the unitbody ground.

Although I would expect a high degree of inherent isolation using an IGBT, how do they avoid feedback issues with the lower voltage side? Unless the ground sides were both bonded together. Otherwise you would have to ensure absolute isolation from the MG's to the inverter/converter and back to the HV pack.

That's why in most residential applications the ground and neutral are bonded, so you don't develop dangerous potential differences between neutral and ground. In most industrial applications, depending on Wye or Delta transformer, the neutral is either kept at a separate "reference" or bonded to ground.

Yes, all my Fluke meters support the optional ampclamp adapter, but it only has resolution down to 1 mA. I did try that in Ready mode but the meter read 0.0.

I did that. Across the bonding points the potential was 1 mVdc. From a common ground reference point (The suggested ground point for boosting, that nut on the firewall passenger side near the brace) I then obtained the readings I posted earlier, 48 mVdc.

Uh, thanks for the reminder, but I do have a B.Sc. and do a lot of process control troubleshooting. The only time I scratch my head is trying to figure out odd +40th order harmonics that seem to drive VFD motor drives bonkers. A lot of capacitive switching in input power will cause those odd harmonics.

I went to a different Toyota dealer here who is having a "grand opening" at a new larger location. I wanted to spend more time poking around under the hood of an Echo without drawing too much attention to myself.

Plus they had milk and cookies, and it's unheard of for me to pass on a free feed.

The Echo Hatchback battery is located at the right side under the hood, I mean the Driver side. Right side looking down at the hood. I was really surprised at how dinky the ground cable is.

It splits from the ground clamp into two dinky wires, one immediately goes to the unitbody as a ground. The other wire appears to snake under the battery and to the motor accessory bracket, but I didn't want to poke around with my flashlight and draw obvious attention to myself.

Those ground cables are *really* tiny, way smaller than the positive cable that goes to the starter. If I had to guess, no more than 8 ga. It looks like it is somewhat thinner than an RG-56 coaxial.

So I wouldn't doubt that using an additional ground would provide some benefit to an Echo sedan and an Echo Hatchback. With the Prius, most of the power is being used by the HV side. And I still haven't heard from anybody if there is any worry of false triggering of the GFCI for the HV side by using an additional ground bond.

 

Would this grounding technique improve any noise problems I may have with putting an inverter and computer in the Prius and connected to the stereo? I have heard that vehicles not properly grounded my have noise problems with pc's and inverters.

 

Maybe, though wait to see if you actually have a problem first.

I know a guy here who does independent computer network and fiber optic installs for large customers and government customers. He has a Ford one ton van with the 7.3 diesel, two alternators, an insulated cube box on the back, and a large Tripp-Lite inverter. I think it's 5,000 watts.

If he has the inverter on, it used to be impossible to listen to the radio even FM as a screeching whine came out of the speakers. He put in a grounding kit, including separate shielded wiring to the auxiliary relay for his inverter. That helped a lot but the noise is still there.

Considering the equipment he is running in his cube van, I don't think it will get any better. A lot of that equipment is Class A (Commercial use) so he has to expect a certain amount of interference. Again though, he made the situation a lot better with his grounding kit.

 

I already have a Coleman 800w inverter. Would just running the inverter off the ACC battery cause the interference? I plan to buy all the parts in a few months and would hate to find out that I should have gone dc-dc for the PC instead of inverter. With the inverter method I will be using a 400w PSU with dc-dc the best I can hope for is 250w.