Recent Posts

31

Microcontrollers / Re: Is there something to learn for embedded/IOT from the Crowdstrike disaster?

« Last post by langwadt on Today at 08:31:12 pm »

Learn to do backups 

Backup is not an issue here whatsoever, no data corruption happened. The issue is about how do you boot many thousands of computers to delete the file that causes BSOD. Basically needs a physical access and manually doing one by one unless it's running on a virtual machine.

and if you are using bitlocker, which is default now I believe, you need the key you can't just boot in safemode 

32

Beginners / Re: Fast snubber circuit for relay coil

« Last post by Zero999 on Today at 08:24:15 pm »

Other than that, yep, it won't be faster than a true open circuit, but the skin effect of the armature itself (who knows how it's constructed, maybe it's solid, maybe it's laminated, but there's a cutoff in there whatever the case) will have a minimum decay in some ~ms, so it probably doesn't matter beyond the zener+diode or S-D clamp zener cases, which grant unlimited survival, and well-defined power dissipation limits.

The OP asked a very simple question: will his suggested circuit work to speed up the relay release time. The answer is:

No, it will slow it down.

The thread has wandered off completely, and is going on about back emf spikes. But that is not what the OP was asking about. Every circuit proposed in this thread will leave the relay release time unaffected or will increase it.

The only two people in this thread to get it right are @langwadt and me.

You might be right in theory, but you're wrong in reality, when the switching transistor is destroyed by the back-EMF spike.

Selecting a snubber is a compromise between switching the relay off quickly and not damaging the switching transistor, hence the reluctance to make blanket statements such as no, it will slow it down.

33

Microcontrollers / Re: Is there something to learn for embedded/IOT from the Crowdstrike disaster?

« Last post by wraper on Today at 08:21:53 pm »

Micro$oft and Crowdstrike will survive this because they have contractually excluded all liability and anyway nobody is big enough to sue them.

MS is not liable whatsoever. They have nothing to do with causing the problem. However MS stepped in and released a tool for simplifying the fix.

34

Projects, Designs, and Technical Stuff / Re: Compact 300V 300mA high side switch with PNP

« Last post by David Hess on Today at 08:21:24 pm »

I don't think it's possible to be as compact with a N Mosfet solution. You need a driver with a 10-20VDC power supply for the bootstrap power supply + HV bootstrap capacitor.

That is why I suggested a photovoltaic optocoupler like the TLP591B which generates its own 10 volt supply.  Only the high voltage n-channel MOSFET and the photovoltaic optocoupler are required, so 2 parts.

35

Microcontrollers / Re: Is there something to learn for embedded/IOT from the Crowdstrike disaster?

« Last post by wraper on Today at 08:17:59 pm »

Learn to do backups 

Backup is not an issue here whatsoever, no data corruption happened. The issue is about how do you boot many thousands of computers to delete the file that causes BSOD. Basically needs a physical access and manually doing one by one unless it's running on a virtual machine.

36

Power/Renewable Energy/EV's / Re: Thyristor gate driver problem

« Last post by artis on Today at 08:16:14 pm »

The unijunction transistor is the actual device which controls the firing angle.
The NPN transistor appears to be what the GE SCR Manual called the ramp and pedestal circuit, to provide higher resolution.

Yes it seems that way to me too. But the more I look at it , it looks like the BC107 transistor is almost useless as all it does is provide a parallel path for the current to charge the 0.68uF cap to fire the UJT. But why the transistor, why not just one potentiometer as current path to the 0.68uF cap? I fail to see how that increases resolution?
I am planning to bypass the bc107 transistor and the 23k resistor and just leave the 60k potentiometer and see how that solves the problem.
The diode bridge creates 100 Hz half periods so each half period as it starts it starts also charging up the 0.68uF cap and if it was done only through one potentiometer then the pot would determine how fast the cap charges which would determine the place in the half period at which the UJT fires the thyristor, but now the same happens only through 2 parallel routes as it seems to me, firstly through the 23k resistor which is fixed and secondly through the 1000uF cap in series with the 60k pot series with the 0.68uF cap, when the 1000uF cap has opened the bc107 bipolar, the charging happens faster.

Anyway I would appreciate your thoughts and comments, is the bc107 close to useless in this schematic and can be omitted?

37

Microcontrollers / Re: Is there something to learn for embedded/IOT from the Crowdstrike disaster?

« Last post by wraper on Today at 08:12:32 pm »

Don't deploy to prod on Friday.

38

Microcontrollers / Re: STM 32F4 FPU registers and main() gotcha

« Last post by dietert1 on Today at 08:04:59 pm »

https://forums.freertos.org/t/cortex-m4-hard-fault-when-using-floating-point-unit/10180
Now that i read it once more, it isn't that clear whether increasing stack helped or using the CPAR register or both.

39

Test Equipment / Re: Difference between HP / Agilent 435A and 435B Power Meters

« Last post by metrologist on Today at 08:04:09 pm »

I think mostly you found the difference.

http://www.test-italy.com/Specifiche%20strumenti%5CHP%5CHP%20435A,11683A%5CHP%20435A.pdf

https://xdevs.com/doc/HP_Agilent_Keysight/HP%20435B%20Operation.pdf

40

Repair / Re: What is this component? - Is this a voltage regulator?

« Last post by fzabkar on Today at 07:54:56 pm »

I'm not confident. In fact I think it's wrong.

AZTP = AZ10EP16VST

https://datasheet.sisoog.com/file/7a657573/datasheet/925f8e38c367f3a59879e012ba3cf9c8553025a9.pdf