PHEV Winter update part 2

Some time ago I quickly mentioned about hybrid Outlander's winter performance. After some more cold weather it's time to make an update. It's been hovering near -30 C for some time now so I think we got nice impression on what to expect. And it's not pretty.

So, like I mentioned, when it got colder, fuel usage went up. Way up. The mask previously mentioned helped a bit, but still nowhere expected figures.

At this point the temperature performance looks like this;

Down to -10 C: Fully electric mode available, with severely reduced range (about 20 km). Heating works too, with electric heaters (I think). Car works fine if not in EV mode, but fuel usage is increased.

Down to -20 C: Electric mode no longer available, and car mainly moves with gasoline. Drive battery isn't used much. Fuel consumption up to previously mentioned 8l/100km figure, possibly due to keeping engine warm at all times.

Below -20 C: Drive battery isn't used at all, it's almost like it weren't there. Fuel consumption way up, around 10l/100km figure.

Below -30 C: Not yet experience this, but I hope it still starts. Come to think of it, it's been several winters now since we really had temperatures below -30 C around here, and I don't even remember when it was down to -40 last time.

So this hybrid tech isn't that great in cold weather. Not that surprising, really, although I have to admit I secretly hoped a bit better performance. Fortunately however "really cold" weather is limited usually to just few weeks in a year, and "cold" to several months, so most of the time I get to enjoy the benefits of (limited) electric drive.

Does your scale lie to you?

I guess almost everyone has a bathroom scale somewhere. I also guess that quite a few people are somewhat nervous if not scared to use theirs (I may or may not be one of them), but that is not the topic here.

These days scales are mostly digital. When you get on the scale, it ponders a while (flashing something on display for several seconds) and then, unerringly, shows you the result. There's something that feels trustworthy about that.

You may try this. Step on your scale several times and take readings. Pretty close to same number every time, right? Even with the cheapes possible scale you can find.

But is it really so?

A friend tried this one time. He got on a scale, and got a reading. Tried again, got same reading. Third time, the same.

Then he told his wife to step on.

And then he tried again himself. This time the reading was off from previous reading by several kilograms.

Curious...

After some more experimenting he determined that the scale was horribly inaccurate, and to cover this it remembered last reading, and if new measurement was relatively close to last one, it showed the previous reading again. Different enough weigh reset this memory and started with new number.

Now, one should not expect cheap(est) bathroom scale to be very accurate anyway. Even in that kind of relatively simple analog-to-digital system, there are many places where errors creep in. Main measurement components, amplification and A/D converter as primary culprits.

And when you are trying to shave every cent from the cost of the scale, these things suffer.

I've said this before, and it's worth repeating here again: Never buy a tool from the very cheapest group (make that from two or three cheapest groups if you're ordering directly from china), as it will be pure junk. Pick one from the next price group up, as it very likely is of much better quality.

Artificial Intelligence

Before I start, understand that I do not work, dabble or have never even played with with any kind of AI/ML (Artificial Intelligence / Machine Learning) systems, so everything that follows is my interpretations based on relatively superficial articles and discussions I've read and may be partially or even completely wrong. You've been warned.

There's been a lot of talk about AI lately, how it is going to be massively change everything, from work to society to ... everything. I am a bit skeptical though, as everything I've read points the current state of AI being mostly hype. Even the term AI is wrong at the moment. There is absolutely no intelligence involved there - only small mathematical models that have been trained to give specific output with various inputs - Machine Learning. Like identifying objects from pictures - is this a cat? chair? elephant?

Don't get me wrong here, making that work is impressively difficult task. But the problem is that this is just about the point where these systems stop. They do that one - and only one - task well, and nothing else. Yet they still are immensely useful.

These things are essentially black boxes however - you feed them training data, telling them what the expected response for that data is, until you get reliably the result from them you want. Which is nice, but if you ask specialists how they work internally, you get no response. They don't know what happens inside, it "just works". Until it doesn't, and you can't really even figure out why it suddenly doesn't work. Say, wouldn't it be fun if your automatic car's traffic sign detection suddenly went haywire just because there's a balloon or something else equally unexpected somewhere in camera view...

I'm using car here as an example a lot, since people reading this are likely to have at least passing understanding of driving as an challenge, even for human mind.

Worse yet, you can't even train these algorithms when they are deployed. Or more specifically, you can't allow them to be trained (or allow them to train themselves) in uncontrolled environment - like when being used in your car, because that might too easily train them to behave in erratic fashion, or simply wrong.

And then we return to issue of these models doing just one job. How do you combine them to produce useful result, even in unexpected situations? And how do you define unexpected in the first place - remember, these models occasionally behave in weird ways when they see data they didn't expect. So what your car's collision avoidance system does when it suddenly sees 5-feet high milkshake in front of it?

All that being said, I have no doubt that in very strictly defined jobs and environments these systems are absolutely wonderful, say, as quality control system picking out bad products in factory or something like that. But considering all things I mentioned above, I dare to say that we're still very, very far away from useful general-use artificial intelligence, and super-minds visioned in both utopian and dystopian sci-fi is currently just that - fiction.

But I still don't rule out the possibility of it getting suddenly much closer due to some new innovation.

Santa Fe key fob fix

The other car I have - Hyundai Santa Fe 2003 - will be replaced this spring or summer, but still I have to do some upkeep. Including clutch replacement which I had to do few months back (cost: about 800€ total, which I will definitely not be getting back in re-sale value).

But other annoyance has been the key fob, specifically the remote control. It has apparently been troublesome in past, too, as previous owner has really mangled it at some point, to a degree it can't be closed without some external aid (he had used duct tape; ugly, ugly..)

Without this fob working the re-sale value of this car would be even lower, so I pretty much had to get it working now, so I opened it up. Battery was good, so that wasn't an issue this time.

As expected, it has single board and battery inside it. Battery here is hidden underneath the rubber part (in middle) and connects to PCB with contacts underneath.

After some probing, I figured out that this time, both buttons had some contact issues and were not toggling properly, and that solder joints on the main chip were bad too. Latter was easy fix; apply some flux, some solder and clean up. Done. Buttons, however, were a bit more troublesome, as they were a bit special - just 1,5mm thick, or 1,2mm (or maybe 1,3? can't remember now) when pressed. Not exactly common part. On picture above the buttons have been removed already (hot air station with controllable temperature and air flow is very useful for this kind of rework)

After some searching I found button that appeared to be identical to originals - but, as usual, this was no longer produced and also out of stock. Damn.

Trying to find suitable component can be sometimes difficult. Especially when you are not exactly sure what you are looking for. This wasn't even worst one to look for - parametric search most sites offer is good enough to narrow parameters down and then it's mostly browsing and checking datasheets for details.

On the other hand, I was just looking for suitable connector. In that case I didn't have any details in mind, just that I want board-to-wire connector where I can very easily change a wire, as target device was new version of test module for a finished product (so bed  of nails-type tester wasn't an option). For testing I need to insert four wires to tester, run test, then disconnect. So quick and easy connect and disconnects were priority.

Opening just "connectors" page of Farnell gives me somewhere around 400 000 (four hundred thousand!) different connectors and connector parts. Fortunately I could quickly ignore some categories completely (Circular connectors, RF connectors, FFCs and so on) but still I ended up with thousands still to browse through.

I wouldn't mind browsing through thousands of connectors if the page loading were fast - which farnell's page isn't. It's sloooow. So slow that I'd actually prefer to browse old-style catalog with thousands of paper pages instead. Sometimes after some browsing I end up with handful of options which I will order some pieces to figure out with physical model with would be the best. Not this time though. In the end I couldn't find better - or even close to it - connector for this job than old spring-loaded speaker connectors I already had though of. So guess I'll settle for them for now for that job.

But back to this fob. After a bit more searching I found almost identical button. Dimensions were almost exactly same, but this didn't have similar SMD contacts than the old ones, but instead just small pads underneath. Not great fit, but close enough - I managed to get them soldered in, with fairly liberal amount of (leaded) solder and after gluing the fob back together (not a prettiest of jobs, but better than duct tape) it's working again.

So at least that's one less issue to think about.