Replacing AC compressor?

@lech auto air conditioning has been incredibly helpful on here over the years... Perhaps he will see this and provide a link... Or do a google search with his username and "priuschat" and you'll find plenty to read.

 

... by which I hope you mean you remembered to remove the orange service plug from the battery in the rear of the car (and kept it in your pocket, as the recommended practice), and allowed enough time for inverter capacitors to discharge.

 

Okay, as I was elbows deep in my battery pack a few days ago I must ask. What capacitors and how much juice are we talking? I remember highschool electronics class and iirc good capacitors can hold a charge for well over a month. (I got my fair share of curiosity kills the cat "what's that on the counter? OW") Obviously I had the orange plug pulled, but in reading the how to's I didn't see that bit. I'm very curious as I expect to be back in the pack next year.
@ChapmanF thank you in advance for the information.

 

Ain't nobody talk about no dangerous capacitors on PriusChat almost ever... Maybe its a conspiracy?

 

That link didn't seem to take, lemmee try: @lech auto air conditionin , without the "g" at the end??

Might be something worthwhile in the attachment, Repair Mainual excerpt on compressor remove/install:

 

The capacitors in question are inside the inverter/converter assembly under the hood, so when they hold charge, that charge appears everywhere on the "frame wire" from its connections at the inverter back to the car-side terminals of the main relays in the battery.

Although a bare capacitor can hold charge for a very long time, it is pretty much standard electronics practice for anything built with a dangerous amount of capacitance to also be built with a permanently-connected resistor to bleed charge off, with a known amount of time to bleed down to a less-dangerous level. Documentation doesn't have to give the exact values of the capacitance or the bleed resistance, but is supposed to tell you the bottom line, how long you should wait for the charge to bleed to a safe level.

In the Gen 2 Repair Manual (more info), in the Introduction section where it explains removal of the service plug, it also says "after removing the service plug, wait 5 minutes before touching any of the high-voltage connectors and terminals. HINT: 5 minutes are required to discharge the high-voltage condenser inside the inverter." ('Condenser' is sometimes used as an older term for 'capacitor'.)

To categorize the risk of a particular capacitive system, you need to know a few things about it: the voltage it is charged to, its capacitance, and the resistance of the bleed resistor. From that you can calculate the time constant for it to bleed off, and you can also rate it on several different kinds of hazard, including electrical shock, thermal hazard through heating an errant wire or tool, hazard of arc flash burns, and arc blast pressure hazards to body tissues or eardrums. There is a standard, "electrical safety in the workplace" (NFPA 70E) with an "informative annex" section on how to calculate those risks.

In the case of the Gen 2 inverter capacitors, we know the voltage they charge to (same as the battery, 201.6 V nominal, can be up to 270 or so in operation*). We're given the five-minute wait-for-discharge time, but we're not given the actual capacitance. Anybody who has had a Gen 2 inverter apart might have seen the printing on the caps. I haven't.

I gave an example of how to work through the NFPA 70E hazard calculations in this post, using a different example, an add-on "plug-out" inverter of 3 kVA capacity, after taking it apart and seeing that its input caps total 1360 µF. In that example, the voltage is also the same as the 201.6-maybe-270 V of the battery, and the risk calculations came out that it can very definitely give you your last heartbeat, but comes out low-risk on the other (thermal, arc flash, arc blast) types of hazard.

Until somebody comes along and says "yeah, I had a Gen 2 inverter apart and saw the caps are ____ µF), I might just assume that because the car's powertrain inverter handles roughly ten times the power (in the 30 kW ballpark rather than 3 kW for my plug-out example), its input caps might also be scaled up some, and that could show up in the hazard calculations if somebody wanted to do 'em.

* there are other capacitors inside the inverter assembly that get charged to the higher 'boost' voltage, up to 500 V in Gen 2 or 650 V in Gen 3, but those aren't the ones that are connected across the battery input. Introduce yourself politely to them if you're ever inside the inverter assembly, though.

 

Yes Prius chat must have a limit on a name size so the G could not fit at the end

 

Yep, that would be me. "Yeah, I have a Gen2 inverter apart and the saw the capacitor is labeled as DC600V 1 130 uF + DC600V 282uF + DC750V 0.1 uF"


Posted via the PriusChat mobile app.

 

Hot damn!!!!! You have no idea how much I appreciate you taking the time to educate me. My expertise is in structural building engineering and haven't done much with electronics in almost 30 years. Your post was awesome!!!

Thank you so much!!! @ChapmanF

 

Thanks! So Panasonic made a big custom potted thing with several of the capacitors used for the 'battery' side bus and the 'boosted' bus.

If only they also said which of those was which.

Enter the Oak Ridge National Lab teardown of the 2010 Prius that they published in March 2011. They helpfully annotated a block schematic and gave a table with the capacitances across various Prius, Camry, and Lexus models:



Here, the cap that goes across the battery connections is the one ORNL labeled "Filter Capacitor", and it turns out that's not the biggest one in the custom potted thing. In the Gen 2 Prius, it's the 282 µF one.

So that one's really not all that big ... smaller than what was in the 3 kVA inverter of my other example, even though this inverter handles ten times the power. When a 282 µF cap is charged up to 270 volts (as can happen), it is holding ½ * 0.000282 * 270² or just over ten joules of stored energy. That voltage and stored energy still falls (just) into the range for significant risk of death from shock and heart fibrillation. My earlier example with the 1360 µF cap scored low on the thermal, flash, and blast risks, and so this scores even lower on those.

Way down at the battery's nominal voltage of 201.6, the energy stored here is just under six joules. That takes it out of that significant-last-heartbeat-risk category, though I don't know that I'd go provoking it just to prove a point. Anyway, it is still enough "to cause serious injuries, either directly or through reflex action" and can "knock a body part into nearby objects, or in some cases cause severe muscle clenching, temporary paralysis, or dislocated joints."

The bigger, 1130 µF cap in that custom block is the one ORNL labeled the "smoothing capacitor". That one's across the inverter assembly's internal boosted bus, and can be charged up to 500 V in Gen 2. That right there represents 141 joules of stored energy and responds well to being called "sir".

 

141 joules... That would "severely chastise" someone who was insufficiently respectful.

Posted via the PriusChat mobile app.

 

The 'thermal hazard' categories (where, for example, you inadvertently short it with a tool you're holding or a ring you're wearing and the tool or ring burns you) start at around 100 J, so yes, 141 J is getting its nose into another category of hazard.

NFPA 70E would also have you wear ear protection when working around anything over 100 J.

For perspective, the arc flash PPE requirements start at 10 kilojoules. That would be more of an industrial- or utility-sized capacitor, about seventy times more stored energy than here.

After ears, the next arc blast hazard is to collapse lungs, and that risk is negligible up to about 112 kJ, nearly a thousand times more than we're talking about here. At that level or above, there's no such thing as recommended PPE; you just have to take the actual numbers and calculate the lung-safe boundary distance, and design things so nobody ever has to go inside it when the capacitor isn't known discharged and shorted.