Test Equipment Anonymous (TEA) group therapy thread

[tggzzz]

When I drove my first automatic the engine stalled at the first traffic light. It took me too many sweaty minutes to realize that you have to put it in neutral before you can start the engine again 

My first car tended to drop into neutral - just after entering a junction and starting to pull away. Eeeek!

The engine oil became dirty, and the same oil was used for the automatic gearbox. Dirt confused the hydraulic logic

To my knowledge, no manufacturer in the history of ever has done this. ever.

Motor oil needs to be changed every 3K miles, and is not safe to use in the gearbox as clutch packs and planetary gears need higher shear-rated oil, plus anti-foaming agents which will burn the rings in an engine. Literally two different families of lubricant. This is why ATF is good for 60-120K miles.

4-stroke motorcycles, I've seen this; but they're just a wet-sump clutch. Different animal altogether.

mnem
 

Well, you've just gained knowledge. It was a 1970s Austin Allegro 1100, a truly dreadful machine.

I'm pretty sure the auto box was only offered in the 1275cc "1300" and larger engined Austin / BLMC cars.

It probably was a 1300.

[BU508A]

Meanwhile on Mars:

Houston, we have a touchdown!

NASA's Perseverance Has Landed

Cheers erupted in mission control at NASA's Jet Propulsion Laboratory as controllers confirmed that NASA's Perseverance rover, with the Ingenuity Mars Helicopter attached to its belly, has touched down safely on Mars. Engineers are analyzing the data flowing back from the spacecraft.

JPL:
https://mars.nasa.gov/mars2020/timeline/landing/status/

[mansaxel]

good reading for SW developers regarding encoding

"Before I get started, I should warn you that if you are one of those rare people who knows about internationalization, you are going to find my entire discussion a little bit oversimplified. "

Yeah. As a multi-facet victim of character set ignorance, I sort of fit that description. My name contains a letter, Å, (Unicode U+00C5) which is known in about 5 languages (which is less than its umlaut neighbours Ä and Ö which are common in for instance German too). It exists in ISO 8859-1, a.k.a LATRINE-1, but since people historically have been extremely bad at doing LATRINE-1 it usually ended up with a truncated 8th bit, turning Å into E. Yay. Or not.

tl;dr: Always use UTF-8. It is bleeping 28 years old now, and is implemented in any OS modern enough to deserve a network connection. (I am aware of the EBCDIC universum, and choose to ignore it.)

</rant>

[bd139]

Meanwhile on Mars:

Houston, we have a touchdown!

NASA's Perseverance Has Landed

Cheers erupted in mission control at NASA's Jet Propulsion Laboratory as controllers confirmed that NASA's Perseverance rover, with the Ingenuity Mars Helicopter attached to its belly, has touched down safely on Mars. Engineers are analyzing the data flowing back from the spacecraft.

JPL:
https://mars.nasa.gov/mars2020/timeline/landing/status/

Glad to hear it had more to say than “thud” like our last attempt with Beagle 2 

[bd139]

good reading for SW developers regarding encoding

"Before I get started, I should warn you that if you are one of those rare people who knows about internationalization, you are going to find my entire discussion a little bit oversimplified. "

Yeah. As a multi-facet victim of character set ignorance, I sort of fit that description. My name contains a letter, Å, (Unicode U+00C5) which is known in about 5 languages (which is less than its umlaut neighbours Ä and Ö which are common in for instance German too). It exists in ISO 8859-1, a.k.a LATRINE-1, but since people historically have been extremely bad at doing LATRINE-1 it usually ended up with a truncated 8th bit, turning Å into E. Yay. Or not.

tl;dr: Always use UTF-8. It is bleeping 28 years old now, and is implemented in any OS modern enough to deserve a network connection. (I am aware of the EBCDIC universum, and choose to ignore it.)

</rant>

Never end up anywhere senior in software. All you do is encoding problems, time zones, synchronisation, threading issues, discover new and interesting failure modes and lose your hair.

[Robert763]

In July 2019 a large (3m long 95kg) UAV took off from Goodwood in the UK, lost control, climbed to 8000' into Gatwick's air space and then crashed.
The accident report was published today. The design and construction of the control system on this thing is truly shocking! Hardly fit for a toy RC car
https://assets.publishing.service.gov.uk/media/602bb22f8fa8f50388f9f000/Alauda_Airspeeder_Mk_II_UAS_reg_na_03-21.pdf

[mansaxel]

Well I found the problem with my internet speed and the issue is me. The old Netgear router isn't up to the task. Doesn't support 200mbps. The most I can jam through it is about 87mbps. If I connect directly to the cable modem and bypass the router I get 230mbps.

I have another older Netgear router and it doesn't support it either. In theory they are both 150mbps routers.

So off to Amazon. The one coming supports 1.2gbps. Ya think that's enough? 

Will be interesting to see how it performs in your real world tests! 

I've always found that even a good router cuts speed by about 20%, just for having to "work" the packets...

Then you have never seen a good router. Mine, a Cisco 7200VXR will happily saturate a 1Gbit link with imix (packet size distribution most often found in Internet traffic) and does not detract from the link speed otherwise; not unless you force it to do funny things.

Of course, as soon as you start doing address translation, and have a shitty DNS proxy in your router, then all bets are off.

I just tried out my connection, which is 100Mbit symmetric Ethernet. (actually, it's 1GE but it's capped to 100Mbit) and from my laptop, over local wifi, I can get 87 Mbit TCP upload to my server, as measured with iperf. And that's through NAT.

I tried from one of my virtual machines which is not NATted and has a 1GE shared connection into the switch and then to the router, and as expected I get slightly better results;

Code: [Select]

[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec   113 MBytes  95.1 Mbits/sec  700             sender
[  5]   0.00-10.00  sec   113 MBytes  95.0 Mbits/sec                  receiver


[Cerebus]

In July 2019 a large (3m long 95kg) UAV took off from Goodwood in the UK, lost control, climbed to 8000' into Gatwick's air space and then crashed.
The accident report was published today. The design and construction of the control system on this thing is truly shocking! Hardly fit for a toy RC car
https://assets.publishing.service.gov.uk/media/602bb22f8fa8f50388f9f000/Alauda_Airspeeder_Mk_II_UAS_reg_na_03-21.pdf

Thanks for that. One normally hears about things like this in the news but never gets to hear the end of the story; the publishing of the enquiry doesn't have the same news impact as the airframe did with the tarmac, unless it's "A minister bribed with a treacle covered floozie to look the other way was the cause of the failure of ...".

[bd139]

Difference between *maker* and *engineer* obvious in that one...

[tggzzz]

Difference between *maker* and *engineer* obvious in that one...

Yup.

As I sometimes point out, makers consider how things work but engineers consider how things fail.

[bd139]

And software wranglers wonder how the hell it ever worked while watching it fail 

[SilverSolder]

Well I found the problem with my internet speed and the issue is me. The old Netgear router isn't up to the task. Doesn't support 200mbps. The most I can jam through it is about 87mbps. If I connect directly to the cable modem and bypass the router I get 230mbps.

I have another older Netgear router and it doesn't support it either. In theory they are both 150mbps routers.

So off to Amazon. The one coming supports 1.2gbps. Ya think that's enough? 

Will be interesting to see how it performs in your real world tests! 

I've always found that even a good router cuts speed by about 20%, just for having to "work" the packets...

Then you have never seen a good router. Mine, a Cisco 7200VXR will happily saturate a 1Gbit link with imix (packet size distribution most often found in Internet traffic) and does not detract from the link speed otherwise; not unless you force it to do funny things.

Of course, as soon as you start doing address translation, and have a shitty DNS proxy in your router, then all bets are off.

I just tried out my connection, which is 100Mbit symmetric Ethernet. (actually, it's 1GE but it's capped to 100Mbit) and from my laptop, over local wifi, I can get 87 Mbit TCP upload to my server, as measured with iperf. And that's through NAT.

I tried from one of my virtual machines which is not NATted and has a 1GE shared connection into the switch and then to the router, and as expected I get slightly better results;

Code: [Select]

[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec   113 MBytes  95.1 Mbits/sec  700             sender
[  5]   0.00-10.00  sec   113 MBytes  95.0 Mbits/sec                  receiver


My WiFi is quite long distance, through several floors, so I can't get much more than about 50Mbits/sec.

I'll have to have a play with iperf and see what the local network is able to do - I hadn't heard about iperf before, looks pretty cool!

[beanflying]

Difference between *maker* and *engineer* obvious in that one...

Difference between a Marketing driven Company built on BS Hype and an Engineering based approach.

[bd139]

Difference between *maker* and *engineer* obvious in that one...

Difference between a Marketing driven Company built on BS Hype and an Engineering based approach.

Adafruit vs Boston dynamics 

[beanflying]

More like the Theranos approach. Spin it. hype it. show some glossy renders and a scaled poorly made prototype for funding from plebs to strap real humans into a flying coffin.

[bd139]

Random telefunken advert from the 1950s:



Showing this to the kids tomorrow. Wonder if they’ll be asking why there are condoms throwing sperm around inside a TV 

[BU508A]

There is a Fluke 8350A (with Nixie tubes!) in the German ebay ads for 100 Euro:

https://www.ebay-kleinanzeigen.de/s-anzeige/fluke-8350a-digital-multimeter-nixie-amateurfunk-messgeraet/1664660248-168-3483



No affiliations with the seller.

[mnementh]

When I drove my first automatic the engine stalled at the first traffic light. It took me too many sweaty minutes to realize that you have to put it in neutral before you can start the engine again 

My first car tended to drop into neutral - just after entering a junction and starting to pull away. Eeeek!

The engine oil became dirty, and the same oil was used for the automatic gearbox. Dirt confused the hydraulic logic

To my knowledge, no manufacturer in the history of ever has done this. ever.

Motor oil needs to be changed every 3K miles, and is not safe to use in the gearbox as clutch packs and planetary gears need higher shear-rated oil, plus anti-foaming agents which will burn the rings in an engine. Literally two different families of lubricant. This is why ATF is good for 60-120K miles.

4-stroke motorcycles, I've seen this; but they're just a wet-sump clutch. Different animal altogether.

mnem
 

Well, you've just gained knowledge. It was a 1970s Austin Allegro 1100, a truly dreadful machine.

Agreed... now I know of two examples of this horrible design that should never have happened; we've got both sides of the bell curve covered.

Why do I get the feeling this was another "automobile" that more qualified as a "motorcycle with a roof stapled on"...? 

mnem
*puttt-puttt-puttta-puttt!!! ~*~*~*~*

[tautech]

Minis weren't too bad unless you had to work on one. 
My ma had a 1275cc GT Clubman and that thing went like stink happily doing the ton which seems mighty fast when your arse is just inches above the road. 
The wife before we married had one, just a plain old 1000cc which were pretty cheap to run but love makes you do dumb things so I had to work on that bloody thing and as they were quite narrow dropped a front wheel in the pit in my garage one day while trying to line it up for some work that I now don't recall. 
Young, dumb, strong and full of cum and a good heave got it back on terra firma while pop unhitched the handbrake and at least he was there for a 2nd attempt to line it up with the pit.

[Specmaster]

The mini wasn't the only cars that had the gearbox mounted in the engine sump and shared the engine oil and also were available with auto transmissions as well, here is a quick list, Mini, Maxi, Metro, Maestro, 1800, and Montego and these came in Austin, Morris, Riley, Wolseley, MG and Vandem Plas variants. I actually had a 4 years a Montego as a company car and here is video about them, very comfortable cars.

[mnementh]

So how many more are we going to add to this list of cars that had a really, really bad design? Showing more examples of a bad idea does not make it a good idea; it just shows there were more than one person who had that bad idea. 

Like that giant-scale UAV built from hobbyist-grade quadcopter parts and being flown over spectators with firmware that clearly didn't have any failsafe programmed.

That said... obviously I was wrong in that nobody ever did it. I stand (well, sit) corrected. Just because i have trouble believing anyone would make such a poor design, does not necessarily make it so.

I guess this proves what my grand-dad used to say... "There is no idea so bad that someone hasn't tried to make a buck off it."

mnem
 

[mnementh]

Difference between *maker* and *engineer* obvious in that one...

Difference between a Marketing driven Company built on BS Hype and an Engineering based approach.

Adafruit vs Boston dynamics 

Hoo boy... there are so many examples of wrong in that UAV I really don't know where to begin. A lot of the hardware I can see is clearly hobbyist-grade parts intended for use on mini & micro racing quadcopters, with no consideration of the exponential increase of the dangers involved in "super-sizing it".

The failsafe as described is almost a "failure guarantee". At one point they stated the relay would kill power to the FC, resulting in no signal to the ESCs which is a huge no-no...   in the event of loss of radio signal, failsafe loops should either attempt to level the craft and slowly descend, or at least kill the craft completely by pulling the ESC signal pins LO and holding them there as long as there is power to the ESCs, which clearly was not implemented in their "custom firmware".

This is why all the popular hobbyist-grade firmwares, while admittedly open-source and written by volunteers, still have multiple avenues of failsafe implementation in the firmware depending on the primary purpose of the copter.  To not take advantage of that as seen here is, IMO, criminally negligent.

The nature of these things makes a relay in the main power nearly impossible to implement, for the same reason a fuse doesn't work. The only viable emergency kill switch is going to be either something that cuts the main battery wire physically, or something that forces the main power connector apart. The latter has been tried, and unfortunately does not always work due to the connectors' tendency to weld shut when moved under high current load.

The problem here, I think... is that the builder was trying to bring the high-angle-of-attack fast-forward-flight performance and responsiveness of our tiny racing quadcopters to this experimental prototype of something almost human-sized. This is pretty much the exact opposite design scope as multirotor craft which are actually designed for flight around people; ie cinematography rigs, inspection drones and firefighting drones. Those are designed for maximum level-flight stability and redundant flight control such that operator control is never lost, or if it is, the UAV lands itself autonomously in a controlled fashion.

I could write aboot what I can see wrong with this thing for hours... but the fact is, most of what is wrong with it is also what is wrong with the mini & micro quadcopter hobby as a whole; only they applied that wrongness to a craft that is exponentially more dangerous due to its size.

Even the little acro ones I like to fly are capable of seriously hurting a person if it comes into contact with soft fleshy bits; they are little flying blenders with four 0.5-1.5HP motors spinning knife blades at 20k-120K RPM. This does not even address the kinetic energy that a 500-800 gram craft can impart if it were to fall uncontrolled from a great height, which any of them can reach if the ESCs just ramp up and stay at WOT.

There's a reason I drifted away from the hobby in general after the FAA jumped all up our butts due to the epidemic of DJI Phantom-flying bozos who kept flying over people and posting videos of their stupidity on boobToob... It's because I realized that with very few exceptions, these things... all of them... planks, helis, quadcopters and drones... are fucking dangerous as all hell, and they're only going to get more dangerous the more crowded this planet becomes and the more people want to play with them and show off their skillzz and shiny new toyyzz.

mnem
 

[Specmaster]

So how many more are we going to add to this list of cars that had a really, really bad design? Showing more examples of a bad idea does not make it a good idea; it just shows there were more than one person who had that bad idea. 

Like that giant-scale UAV built from hobbyist-grade quadcopter parts and being flown over spectators with firmware that clearly didn't have any failsafe programmed.

That said... obviously I was wrong in that nobody ever did it. I stand (well, sit) corrected.

I guess this proves what my grand-dad used to say... "There is no idea so bad that someone hasn't tried to make a buck off it."

mnem
 

Actually it wasn't a shit idea, it really worked very well and was well known in racing circles and there was a race version produced, Mini Cooper S which I believe won lots of races, it was a good rally car and none of the cars that used the gearbox in the sump approach, driving the front wheels had a bad reputation.

[mnementh]

So how many more are we going to add to this list of cars that had a really, really bad design? Showing more examples of a bad idea does not make it a good idea; it just shows there were more than one person who had that bad idea. 

Like that giant-scale UAV built from hobbyist-grade quadcopter parts and being flown over spectators with firmware that clearly didn't have any failsafe programmed.

That said... obviously I was wrong in that nobody ever did it. I stand (well, sit) corrected.

I guess this proves what my grand-dad used to say... "There is no idea so bad that someone hasn't tried to make a buck off it."

mnem
 

Actually it wasn't a shit idea, it really worked very well and was well known in racing circles and there was a race version produced, Mini Cooper S which I believe won lots of races, it was a good rally car and none of the cars that used the gearbox in the sump approach, driving the front wheels had a bad reputation.

No, it really is a bad idea, for the reasons I stated. The tolerances in an automatic transmission are an order of magnitude finer than a manual transmission, or even most of the engine for that matter. Just because a few manufacturers did it does not make it a good idea. It just means they tried a bad idea and it didn't destroy the brand.

Also, people do a lot of stuff on race cars that simply will not hold up long term in a daily driver.

mnem

[Specmaster]

So how many more are we going to add to this list of cars that had a really, really bad design? Showing more examples of a bad idea does not make it a good idea; it just shows there were more than one person who had that bad idea. 

Like that giant-scale UAV built from hobbyist-grade quadcopter parts and being flown over spectators with firmware that clearly didn't have any failsafe programmed.

That said... obviously I was wrong in that nobody ever did it. I stand (well, sit) corrected.

I guess this proves what my grand-dad used to say... "There is no idea so bad that someone hasn't tried to make a buck off it."

mnem
 

Actually it wasn't a shit idea, it really worked very well and was well known in racing circles and there was a race version produced, Mini Cooper S which I believe won lots of races, it was a good rally car and none of the cars that used the gearbox in the sump approach, driving the front wheels had a bad reputation.

No, it really is a bad idea, for the reasons I stated. The tolerances in an automatic transmission are an order of magnitude finer than a manual transmission, or even most of the engine for that matter. Just because a few manufacturers did it does not make it a good idea. It just means they tried a bad idea and it didn't destroy the brand.

Also, people do a lot of stuff on race cars that simply will not hold up long term in a daily driver.

mnem

All I can tell you is that the idea was around for a long time with that series of cars, and they did chalk up a lot of racing wins and rally wins and were used in many parts of the globe. In New Zealand where the 1800 series of cars, commonly called land crab because of their appearance, were known as Kimberley's and were fitted with a specially designed 6-cylinder engine.