Paul and Sabrina's Cheap 3 Phase Inverter (AC Controller) with Field Oriented Control

[thingstodo]

Originally Posted by ale0502

The male and female connectors are 100% compatible? I mean, are well matched?

I believe so. They fit well and the polarized plug fits and holds.

[thingstodo]

Update for Jan 13

I need to change how I do things for testing the AC controller, and things in general .. I guess. I am avoiding editing video, and I don't want to get way ahead with testing without getting video and results posted .. so I'm *NOT* doing what I enjoy - testing and making stuff work - because I'm avoiding the part I don't enjoy - editing the video. At least, I don't APPEAR to enjoy it .. when I *HAVE* to do it?

My apologies for the long post. This is my thought process and what I came up with .. skip to the last paragraph if you are easily bored! Or even hard to bore ..

My wife asked me why I'm not testing my AC controller any more. After all - half my garage is dedicated to the testing .. and her car is out in the cold instead of inside the heated garage so that I CAN do this testing ...

I explained that I was trying to figure out this new-to-me video editing software, Lightworks 12.5, so that I could show a video of the voltage and current on the battery pack, and voltage and current for the DC motor (that video didn't record .. but I still want to know how to do it), and have that synchronize with the all of the stuff that is logged by Paul's AC motor controller ...

She asked why I was doing all of this ... she is aware that most of my videos have under 10 views and she has ALWAYS asked me why I bother posting video at all. That's actually a good question. Why am I editing video when I get so frustrated by that? PLUS, I have been avoiding working on the videos since I DO find it frustrating.

Well, I mostly do videos for me, and for Paul to see the testing so that he can critique or make suggestions. Having everyone else who is interested watch was a no-brainer. It's more effort to restrict who watches the videos than to just make them public.

I discovered Google live. Google live let me show meters, or a couple of videos of meters, so that those measurements were synchronized to the testing. It would be best if one data collection system gathered all of the data for the battery pack, the AC controller, the DC motor ... the temperatures .. and stored it in a Excel file. But I don't have a system that will gather all of that data. So Google live let me record the data so that I could go back after and read a few points at a few speeds where I sort of stopped changing things quickly ... and get an idea of what was going on. Maybe enter some of those points in the spreadsheet along with the data that Paul logs.

Why not use Google live? ... my internet connection is not fast enough to use Google Live reliably. And when Google live does NOT work, you don't find out until AFTER you are all done and now you have no data and no video. But do I NEED what Google live does?

Back to that later ... so far I found out that I can show ONE video as picture in picture, so I can show one pair of meters (voltage and current for the battery pack, for example) as a window that hovers over the Gopro view of my testing. But that does not show me the DC Motor voltage and current ...

One solution (and it may not be the easiest or the quickest) .. I did a (short) video combining a battery pack meter video and a gopro video, side by side ... like I would if it were the Pack voltage and current beside the DC motor voltage and current ...

Then I added that finished video as picture in picture to a side view of the testing and it works ... kind of. I'm not happy with it. Since I have edited the volt meter and ammeter video a couple of times, the video compression used by the editing software is starting to drop updates on the meter video - which is the most important part of the video. A couple of more tries with different software settings and it seems that the size of the output file determines how good the synchronization stays and how much information is dropped from the meter videos.

To summarize - for this solution I think I need to take the two meter videos, combine them side-by-side and clip them to show the meter values only, output that side-by-side video to a new video clip. Then take that new video clip and include it with the Gopro or side view (phone camera) video, depending on which view gets me out of the way so you can see what is going on, and add the audio from the android app that records bluetooth headset audio. At that point I get something that kinda gives me the same thing as Google Live, but I have more control over the placement of the meter videos and how clear they are. After my learning curve, I'm still looking at hours of editing for under an hour of actual DOING a test, plus the hour or so of setup for water cooling, setting up the meters, getting everything turned on and synchronized, then the cleanup after the testing is done, and getting my car back into the garage afterward.

Back to my wife's question about 'Why are you doing all of this' since I have fun with the testing and doing part. There is always some overhead to documentation ... but this seems a bit over the top. AND I'm AVOIDING it instead of HAVING FUN DOING IT!

There IS some POSITIVE from going through this process. You only do a test ONCE so getting video helps to see stuff afterward that I don't see first time around. AND you guys see stuff that I don't see. More eyes catch more possible errors, or procedural problems, or help explain what happened ..etc .. or .. maybe sometimes you just laugh?

SO .. UMM .. I think I'll keep doing multiple cameras, and making a big noise (like a clapper) to make it pretty easy to synchronize the audio and video. I'll try reading values off the AC controller terminal display on the audio track (time code 12, 13, 14) and then review the video of the meters and write down what the voltage and current was for the battery pack, the DC motor, the temperatures, the hand-checked speeds ... maybe that's enough data. I expect that saving the time I would have used WATCHING each video 5 times and scribbling notes on what was wrong, what to look up, what to edit out ... I could use that time to synchronize the video and audio from 5 sources, write down the numbers, and put them into a spreadsheet and have a LOT of time left over. It will only be a dozen or so readings instead of hundreds that you could get from the video (but I'm sure that no one has), but that should do. My test equipment and methods are NOT very accurate anyway! If anyone is still awake, and has opinions on how I can make this process less labor intensive, please comment!

[e*clipse]

I really appreciate your testing.

Perhaps you could use a data acquisition system? I know this sounds like crazy high-end stuff, but I've been truly blown away by how easy and cheap the stuff is.

I recently purchased a couple DATAQ instruments devices for some testing I'm doing on my electric snow cat. You can log all sorts of stuff simultaneously; for example, I measured current and voltage supplied to a pair of DC motors. I'm not the most patient person for getting software running, but their free software worked great with little futzing around.

For example, the DI-149 allows up to 8 analog input channels +/-150V. The instrument plugs into a USB port and runs with windoze software. I think between your videos and some synchronized direct data gathering, you'd have an unbeatable combo.

- E*clipse

[MPaulHolmes]

I say only do the fun part!!!! Priorities for this:
Have fun
Save mankind from lung lesions lol

I agree with your wife! If yiu are getting bogged down in tedious video editing and you are not one of those psychos that actually like that sort of thing, then film if you want, but dont edit a dang thing! Just say "darn it Jim, I'm an engineer, not a video editing person! In fact Jim, I don't even know the official name of a video editing person!"

[P-hack]

I think the clap method is more than anyone could ask for as long as the time index is accurate, should be good enough to piece it together.

FYI, speaking of video testing, just picked up a $200 seek xr (android), it is a little picky about android models (needs OTG with power, 4.3+), works on the kids onetouch out of the box but not my xperia z1s without adding power.
went with it over the flir because it has a much wider temperature range.

[thingstodo]

Originally Posted by e*clipse

I really appreciate your testing.

Perhaps you could use a data acquisition system? I know this sounds like crazy high-end stuff, but I've been truly blown away by how easy and cheap the stuff is.

I recently purchased a couple DATAQ instruments devices for some testing I'm doing on my electric snow cat. You can log all sorts of stuff simultaneously; for example, I measured current and voltage supplied to a pair of DC motors. I'm not the most patient person for getting software running, but their free software worked great with little futzing around.

For example, the DI-149 allows up to 8 analog input channels +/-150V. The instrument plugs into a USB port and runs with windoze software. I think between your videos and some synchronized direct data gathering, you'd have an unbeatable combo.

- E*clipse

Hmm. Is this what you are describing? DI-149 USB Data Acquisition Starter Kit

What I get from the description is that it will read +/- 10V but will withstand +/- 150V for a short period or +/- 75V continuous. Or am I looking at some old data?

[freebeard]

Not having seen the layout in your garage, this is speculation, but...

Place a large mirror at 45° to the gauges. Place a small mirror to occlude half (or 1/4th) the field of view in front of the camera. Place the camera so that gauges are visible.

Everything will sync at the speed of light.

[thingstodo]

Originally Posted by P-hack

I think the clap method is more than anyone could ask for as long as the time index is accurate, should be good enough to piece it together.

I hope that is enough data to figure out if something goes wrong

FYI, speaking of video testing, just picked up a $200 seek xr (android), it is a little picky about android models (needs OTG with power, 4.3+), works on the kids onetouch out of the box but not my xperia z1s without adding power.
went with it over the flir because it has a much wider temperature range.

Is that an infra-red scan of your cat?

Seek XR ... Hmm .. Does it give you an actual temperature chart for the colors or are the colors a certain number of degrees apart? Maybe a comparision of relative heat?

[e*clipse]

Originally Posted by thingstodo

Hmm. Is this what you are describing? DI-149 USB Data Acquisition Starter Kit

What I get from the description is that it will read +/- 10V but will withstand +/- 150V for a short period or +/- 75V continuous. Or am I looking at some old data?

Hmm - yea - I was just reading the #'s on the DI-149 I have. I try to keep all the voltages down in the +/-5V range, just 'cause the microcontroller works that way. Most things like temp sensors, pressure sensors, etc etc work in the 5V range. The software makes calibration really easy, so reading the output of the current sensor and graphing it as +/- 300A vs 0>5V is easy to set up.

For something like battery voltage, a simple voltage divider will work to bring it down to an acceptable level for testing.

- E*clipse

[e*clipse]

Originally Posted by P-hack

I think the clap method is more than anyone could ask for as long as the time index is accurate, should be good enough to piece it together.

FYI, speaking of video testing, just picked up a $200 seek xr (android), it is a little picky about android models (needs OTG with power, 4.3+), works on the kids onetouch out of the box but not my xperia z1s without adding power.
went with it over the flir because it has a much wider temperature range.

That is really cool! That would REALLY help for my motor cooling efforts. Do you have more info or some links about those cameras?

- Thanks a bunch!
E*clipse