Author Topic: Somebody goofed up  (Read 1749 times)

0 Members and 1 Guest are viewing this topic.

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Somebody goofed up
« on: September 05, 2019, 06:31:16 pm »


Or not? It's in a very popular dev board. Suddenly its magic smoke escaped today...

Even when the experts all agree, they may well be mistaken.
 
The following users thanked this post: bd139

Offline bd139

  • Super Contributor
  • ***
  • Posts: 11888
  • Country: gb
Re: Somebody goofed up
« Reply #1 on: September 05, 2019, 06:35:06 pm »
That sort of shit is why humans invented QA  :palm:
 

Offline martinayotte

  • Regular Contributor
  • *
  • Posts: 61
Re: Somebody goofed up
« Reply #2 on: September 05, 2019, 06:42:29 pm »
 :-DD
 

Offline fourfathom

  • Regular Contributor
  • *
  • Posts: 235
  • Country: us
Re: Somebody goofed up
« Reply #3 on: September 05, 2019, 06:47:02 pm »
What is "LS1"?  It looks like the problem is there, and not the upside-down transistor.  I've seen people deliberately swap emitter and collector in order to get a slightly smaller saturation voltage (less gain, though).  That by itself shouldn't cause any smoke.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 4774
  • Country: gb
Re: Somebody goofed up
« Reply #4 on: September 05, 2019, 06:52:18 pm »
What is "LS1"?  It looks like the problem is there, and not the upside-down transistor.  I've seen people deliberately swap emitter and collector in order to get a slightly smaller saturation voltage (less gain, though).  That by itself shouldn't cause any smoke.

It looks like a magnetic audio sounder (it would have to provide DC continuity for the circuit to 'work'). The GPIO stuck high would probably do it.
Chris

"Victor Meldrew, the Crimson Avenger!"
 

Offline bd139

  • Super Contributor
  • ***
  • Posts: 11888
  • Country: gb
Re: Somebody goofed up
« Reply #5 on: September 05, 2019, 07:04:10 pm »
What is "LS1"?  It looks like the problem is there, and not the upside-down transistor.  I've seen people deliberately swap emitter and collector in order to get a slightly smaller saturation voltage (less gain, though).  That by itself shouldn't cause any smoke.

Fair point. Gain is close to unity doing this so I'm not sure what the deal is. Perhaps it was in the linear region too. Pd off the scale.
 

Offline magic

  • Super Contributor
  • ***
  • Posts: 1122
  • Country: pl
Re: Somebody goofed up
« Reply #6 on: September 05, 2019, 07:36:29 pm »
If LS1 is something inductive it may easily avalanche the BE junction. People say a lot of bad things about avalanching BE junctions, I don't know how true they are :-//
 

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #7 on: September 05, 2019, 07:54:54 pm »
That NPN has become a short circuit, and as the thing is powered by a LiPo, the LiPo went smoking hot as well. I was trying different "beeps" to find out which one seemed louder/could be heard from farther away, gpio 26 was outputting a square wave @2.8 kHz when it happened...
Even when the experts all agree, they may well be mistaken.
 

Online thm_w

  • Super Contributor
  • ***
  • Posts: 1490
  • Country: ca
Re: Somebody goofed up
« Reply #8 on: September 05, 2019, 09:30:50 pm »
DET402-G-1 attached, seems like the top cover is missing in OPs photo for some reason.
And it does appear to be coil based, which would give a voltage spike.
« Last Edit: September 05, 2019, 09:33:16 pm by thm_w »
 
The following users thanked this post: GeorgeOfTheJungle

Online rsjsouza

  • Super Contributor
  • ***
  • Posts: 3460
  • Country: us
  • Eternally curious
    • Vbe - vídeo blog eletrônico
Re: Somebody goofed up
« Reply #9 on: September 05, 2019, 09:36:59 pm »
What is "LS1"?  It looks like the problem is there, and not the upside-down transistor.  I've seen people deliberately swap emitter and collector in order to get a slightly smaller saturation voltage (less gain, though).  That by itself shouldn't cause any smoke.
A.k.a. common collector configuration.
https://en.wikipedia.org/wiki/Common_collector
 
Vbe - vídeo blog eletrônico http://videos.vbeletronico.com

Oh, the "whys" of the datasheets... The information is there not to be an axiomatic truth, but instead each speck of data must be slowly inhaled while carefully performing a deep search inside oneself to find the true metaphysical sense...
 

Online Circlotron

  • Super Contributor
  • ***
  • Posts: 1632
  • Country: au
Re: Somebody goofed up
« Reply #10 on: September 05, 2019, 09:52:18 pm »
What is "LS1"?  It looks like the problem is there, and not the upside-down transistor.  I've seen people deliberately swap emitter and collector in order to get a slightly smaller saturation voltage (less gain, though).  That by itself shouldn't cause any smoke.
A.k.a. common collector configuration.
https://en.wikipedia.org/wiki/Common_collector
If that’s the case, the power supply rails are reverse polarity.
 

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #11 on: September 05, 2019, 10:00:18 pm »
Now that I think about it, I was flipping gpio every 350 µs, but that's not 2.8 kHz, it's 1.4 kHz. Since it's close to the resonance frequency of the DET402-G-1, could that have accelerated the unfortunate events?

Code: [Select]
: invert 0= if 1 else 0 then ;

26 output: speaker
: beep
    begin
        speaker speaker digr invert digw
        350 delay.us
        key?
    until
    speaker 0 digw
;

Even when the experts all agree, they may well be mistaken.
 

Offline fourfathom

  • Regular Contributor
  • *
  • Posts: 235
  • Country: us
Re: Somebody goofed up
« Reply #12 on: September 05, 2019, 10:20:37 pm »
What is "LS1"?  It looks like the problem is there, and not the upside-down transistor.  I've seen people deliberately swap emitter and collector in order to get a slightly smaller saturation voltage (less gain, though).  That by itself shouldn't cause any smoke.
A.k.a. common collector configuration.
https://en.wikipedia.org/wiki/Common_collector

Not the same thing at all.  Look again.
 

Offline fourfathom

  • Regular Contributor
  • *
  • Posts: 235
  • Country: us
Re: Somebody goofed up
« Reply #13 on: September 05, 2019, 10:26:00 pm »
Now that I think about it, I was flipping gpio every 350 µs, but that's not 2.8 kHz, it's 1.4 kHz. Since it's close to the resonance frequency of the DET402-G-1, could that have accelerated the unfortunate events?

Code: [Select]
: invert 0= if 1 else 0 then ;

26 output: speaker
: beep
    begin
        speaker speaker digr invert digw
        350 delay.us
        key?
    until
    speaker 0 digw
;
Quite possibly. Depending on Q, the voltage ringing might well be high enough voltage to kill the transistor, especially in the upside-down config.  This could be a problem at any frequency. But driving it at resonance could make it worse.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 3517
  • Country: fr
Re: Somebody goofed up
« Reply #14 on: September 05, 2019, 10:30:43 pm »
Oh well, do NPNs not have a right to be used inverted? Maybe some of them just feel "right" when used this way. :-DD
 
The following users thanked this post: GeorgeOfTheJungle, nugglix

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 5692
  • Country: nl
Re: Somebody goofed up
« Reply #15 on: September 05, 2019, 10:43:06 pm »
 
The following users thanked this post: GeorgeOfTheJungle

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 3517
  • Country: fr
Re: Somebody goofed up
« Reply #16 on: September 05, 2019, 10:47:02 pm »
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 8052
Re: Somebody goofed up
« Reply #17 on: September 05, 2019, 10:57:33 pm »
IMHO its just a shitty design.  If the transistor can't stand the max. current the sounder will pass @3.3V DC, they should have included a series resistor to limit the continuous current, possibly with a  capacitor across it so you can get more drive when you pulse it.   Also relying on reverse E-B zenering to turn the transistor on and clamp the back-EMF from the inductive load is just plain nasty.

Alternatively they could have used a NPN+ PNP pair as a complimentary emitter follower, and a capacitor in series with the sounder - that's particularly robust because the non-active transistor will turn on enough to clamp the output if the load tries to drags it more than one Vbe outside the rails.
 

Offline Sal Ammoniac

  • Frequent Contributor
  • **
  • Posts: 909
  • Country: us
    • Embedded Tales Blog
Re: Somebody goofed up
« Reply #18 on: September 05, 2019, 11:09:30 pm »
Now that I think about it, I was flipping gpio every 350 µs, but that's not 2.8 kHz, it's 1.4 kHz. Since it's close to the resonance frequency of the DET402-G-1, could that have accelerated the unfortunate events?

Code: [Select]
: invert 0= if 1 else 0 then ;

26 output: speaker
: beep
    begin
        speaker speaker digr invert digw
        350 delay.us
        key?
    until
    speaker 0 digw
;



There's your problem right there... Forth killed it. When the transistor saw it was being controlled by Forth, it did the only thing it could: commit suicide.  :-DD
« Last Edit: September 06, 2019, 03:00:21 pm by Sal Ammoniac »
 
The following users thanked this post: GeorgeOfTheJungle, nugglix, bd139

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 3517
  • Country: fr
Re: Somebody goofed up
« Reply #19 on: September 05, 2019, 11:14:42 pm »
There's reason in controlling an inverted transistor with an inverted language. ;D
 
The following users thanked this post: GeorgeOfTheJungle, Ian.M, RoGeorge, techman-001

Offline bd139

  • Super Contributor
  • ***
  • Posts: 11888
  • Country: gb
Re: Somebody goofed up
« Reply #20 on: September 06, 2019, 07:22:27 am »
There's your problem right there... Forth killed it.

I actually just spilled coffee everywhere laughing at that  :-DD
 

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #21 on: September 06, 2019, 07:30:21 am »
Where's techman-001 ?
Even when the experts all agree, they may well be mistaken.
 

Online techman-001

  • Frequent Contributor
  • **
  • Posts: 501
  • Country: au
  • Electronics technician for the last 47 years
Re: Somebody goofed up
« Reply #22 on: October 09, 2019, 09:03:31 am »
Where's techman-001 ?

Just stumbling across this topic right now :)

I think the inductive load and possibly the on/off frequency killed the transistor being used in "uncommon emitter mode".

I think it may have also killed it if it was in common emitter mode.

George is that Forth your "ATlast" ?

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #23 on: October 09, 2019, 09:12:16 am »
George is that Forth your "ATlast" ?

Yes it is. The only other Forth I've ever had (and used) is one for the Apple II that John Drapper (a.k.a. Captain Crunch) in person sold me at the NCC in NYC in 1979.

Edit:
But that code above won't run in yours because I've added some words to mine: e.g. delay.us (delayMicroseconds), digr (digitalRead) and digw IIANM.
« Last Edit: October 09, 2019, 09:15:44 am by GeorgeOfTheJungle »
Even when the experts all agree, they may well be mistaken.
 

Online techman-001

  • Frequent Contributor
  • **
  • Posts: 501
  • Country: au
  • Electronics technician for the last 47 years
Re: Somebody goofed up
« Reply #24 on: October 09, 2019, 10:39:34 am »
George is that Forth your "ATlast" ?

Yes it is. The only other Forth I've ever had (and used) is one for the Apple II that John Drapper (a.k.a. Captain Crunch) in person sold me at the NCC in NYC in 1979.

Edit:
But that code above won't run in yours because I've added some words to mine: e.g. delay.us (delayMicroseconds), digr (digitalRead) and digw IIANM.

Just having bought a Apple II from Captain Crunch makes you a legend!

Yes I could see your extra Words.

Forth people never copy each others code, for one we don't need to, and 2) our systems are all different.

Your code is:
: invert 0= if 1 else 0 then ;

26 output: speaker
: beep
    begin
        speaker speaker digr invert digw
        350 delay.us
        key?
    until
    speaker 0 digw
;

Mine would be something like:
( not inverts all bits)

 : beep
    begin
           speaker dup digr not digw
           350 delay.us
           key?
   until
   speaker 0 digw
;

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #25 on: October 09, 2019, 10:54:03 am »
Just having bought a Apple II from Captain Crunch makes you a legend!

Bought him the Forth not the Apple II!

Quote
Yes I could see your extra Words.

Forth people never copy each others code, for one we don't need to, and 2) our systems are all different.

Your code is:
: invert 0= if 1 else 0 then ;

26 output: speaker
: beep
    begin
        speaker speaker digr invert digw
        350 delay.us
        key?
    until
    speaker 0 digw
;

Mine would be something like:
( not inverts all bits)

 : beep
    begin
           speaker dup digr not digw
           350 delay.us
           key?
   until
   speaker 0 digw
;

"not"? Hahaha. Thanks! I'm a beginner, my Forth-fu betters every day, albeit slowly.
« Last Edit: October 09, 2019, 12:48:33 pm by GeorgeOfTheJungle »
Even when the experts all agree, they may well be mistaken.
 

Online techman-001

  • Frequent Contributor
  • **
  • Posts: 501
  • Country: au
  • Electronics technician for the last 47 years
Re: Somebody goofed up
« Reply #26 on: October 09, 2019, 11:35:32 am »
Just having bought a Apple II from Captain Crunch makes you a legend!

Bought him the Forth not the Apple II!

Quote
Yes I could see your extra Words.

Forth people never copy each others code, for one we don't need to, and 2) our systems are all different.

Your code is:
: invert 0= if 1 else 0 then ;

26 output: speaker
: beep
    begin
        speaker speaker digr invert digw
        350 delay.us
        key?
    until
    speaker 0 digw
;

Mine would be something like:
( not inverts all bits)

 : beep
    begin
           speaker dup digr not digw
           350 delay.us
           key?
   until
   speaker 0 digw
;

not? Hahaha. Thanks! I'm a beginner, my Forth-fu betters every day, albeit slowly.

Buying anything from Captain Crunch makes you a legend :)

I'm a Forth beginner also. Forth is easy to learn, but harder to master.

 "not" is part of my default dictionary, but ": invert 0= if 1 else 0 then ; " works perfectly. I mention it because Atlast has "not" in its default dictionary also :)

Oddly the Word "words" only lists part of the actual Wordlist, this is unusual however the "Atlast Primitives: Alphabetical Reference" in the 64 bit html has a complete list of Words, and it is extensive! The Word "WORDSUNUSED" does list all unused Words.  These methods are foreign to me, but certainly not wrong, just different to what I'm used to.

To check if a "not" Word is available: the number returned is the memory location of the Word itself
-> ' not .
34378764744 -> words

This fails because there is no such word as xanadu
-> ' xanadu .
 'XANADU' undefined


Atlast includes these cool packages:
 The REAL package. Enables floating point operations. Primitives: (FLIT), F+, F-, F*, F/, FMIN, FMAX, FNEGATE, FABS, F=, F<>, F>, F<, F>=, F<=, F., FLOAT, FIX.

The SHORTCUTA package. Enables shortcut integer arithmetic operations. Primitives: 1+, 2+, 1-, 2-, 2*, 2/.

The SHORTCUTC package. Enables shortcut integer comparison operations. Primitives: 0=, 0<>, 0<, 0>.

The STRING package. Enables string operations. Primitives: (STRLIT), STRING, STRCPY, S!, STRCAT, S+, STRLEN, STRCMP, STRCHAR, SUBSTR, COMPARE, STRFORM, STRINT, STRREAL. If the REAL package is also enabled, the FSTRFORM primitive is available, as well.

The SYSTEM package. Enables submission of commands in strings to the operating system for execution. This package may be enabled only if the implementation of C used to build Atlast provides the system() function. Primitives: SYSTEM.

The TRACE package. Enables runtime word execution trace. Primitives: TRACE.

The WALKBACK package. Enables the walkback through nested invocation of words when an error is detected at runtime. Primitives: WALKBACK.

The WORDSUSED package. Enables the collection of information on which words are used and not used by a program, and the primitives that list words used and words not used. This facility allows you to determine, in the development phase of an Atlast application, which packages are needed and which can be safely dispensed with. Primitives: WORDSUSED, WORDSUNUSED.


I'm very impressed, I may consider using Atlast on this PC for some X86 programs.

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #27 on: October 09, 2019, 12:47:26 pm »
Oddly the Word "words" only lists part of the actual Wordlist, this is unusual however the "Atlast Primitives: Alphabetical Reference" in the 64 bit html has a complete list of Words, and it is extensive! The Word "WORDSUNUSED" does list all unused Words.  These methods are foreign to me, but certainly not wrong, just different to what I'm used to.

I think I noticed that and fixed it. Not sure if the fix is in the gitlab repo or not, though. It's an easy fix anyway.

Quote
I'm very impressed, I may consider using Atlast on this PC for some X86 programs.

LOL. Good! I'm glad to know. This man is (or was?) the owner of AutoDesk. He wrote most of AutoCAD if I'm not mistaken.
Even when the experts all agree, they may well be mistaken.
 

Offline PCB.Wiz

  • Frequent Contributor
  • **
  • Posts: 379
  • Country: au
Re: Somebody goofed up
« Reply #28 on: October 09, 2019, 10:18:21 pm »

Somebody goofed up... Or not?

Reverse connect is not common, as already mentioned. It does work : it gives lower HFE, and lower VCE, but can give lower VCEsat (at low currents)
Not usually seen when driving a power load.
Did you check the actual device pinout, to verify it actually is wired up 'reversed' ?

The posted link says 120mA, but does not mention the resistance or inductance.

Did you scope the transistor & measure the resistance of DET402 ?
With no flyback clamp diodes, the little npn is doing both the drive and the clamping.

It's demise is trying to tell you, that's too much, on a continuous basis... :)

 
The following users thanked this post: GeorgeOfTheJungle

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • Posts: 1973
  • Country: pl
Re: Somebody goofed up
« Reply #29 on: October 10, 2019, 08:07:24 am »
Did you check the actual device pinout, to verify it actually is wired up 'reversed' ?

Yes, it was connected as in the schematics.

Quote
The posted link says 120mA, but does not mention the resistance or inductance.

Did you scope the transistor & measure the resistance of DET402 ?
With no flyback clamp diodes, the little npn is doing both the drive and the clamping.

It's demise is trying to tell you, that's too much, on a continuous basis... :)

I haven't checked anything else other than its upside-down-ness.

Wonder why they use these tiny fleas that are nowadays all over the place, lol. What's with a good old proven 2n3055 that would still be alive? Or even a bd139? :)

Other than that, the thing is a beauty:



Just delete the buzzer, because if you use it the thing catches fire! (LITERALLY)
« Last Edit: October 10, 2019, 10:05:16 am by GeorgeOfTheJungle »
Even when the experts all agree, they may well be mistaken.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf