another dimmer smoked today outside thank goodness!

On Monday, November 23, 2020 at 3:50:47 PM UTC-5, Chuck wrote:


A device like that needs a proper heatsink.   Even if the metal tab is soldered to the PC it's still inadequate for high power applications.

On 11/23/20 4:09 PM, snipped-for-privacy@gmail.com wrote:


I think that's why they are burning up, then.  Nothing else on the board  
was fried, only the one Mosfet.

On 11/23/20 4:13 PM, Chuck wrote:

ng


 in


the

ut
d
/



tal tab is  


 board  


Going back to my ham radio days, my 2 meter amps of 100 W had heatsinks  
of probably 4x6 and an inch or two thick, IIRC.  Sometimes even a fan  
too.  My guess is that would at the very least be needed for these  
dimmers and even that heatsink size will get it no where near say 300  
W+.  Shame on them.

non inductor ???



For 70A mosfet 7A is light load.  Estimate based on typical
values shows that it should disspate about 0.3W.  For TO-220
part at that load heatsink makes little economic sense: mosfet
will run fine without heatsink and better (lower Rdson)
mosfet is cheaper than heatsink.  Of course, if one wants
to handle nominal 30A, than heatsink would be necessary.

The above assumed that there is an inductor and that second
HY1707 works as synchronous rectifier.  Otherwise RMS current
may be much larger than average.

  snipped-for-privacy@math.uni.wroc.pl wrote:

=============================

** Some uniwanker dope  



** It will be much higher in practice.  




**  The OP's rms current is way higher than 7A.  



** You on crack ???  

Do you have no idea what this device is ?  

It is **NOT** a power supply .

It is just an on/off  switch performing PWM lamp control.  



.....   Phil  

Well, I would prefer to put inductor in such device.  But
you are right, inductor (+ good control) would be too expensive
for current market.  At first I was mislead by mention of two
mosfets, but connecting them in parallel (to get higher current)
in simple PWM makes sense.

On 2020/11/23 12:50 p.m., Chuck wrote:

  

 did  

good  

n  


hers  

e  


urnout  

stead  


tage/  

rust  


No UL approval I'm sure.

Why would you trust it then?

Unregulated electrical junk sold on Amazon can be hazardous to your  
house or your family.

John :-#(#

 John Robertson wrote:
==================
 >  


** What standard is there for such a device the runs on  12- 24V DC ??  



** UL does not check for good design or reliability.  




**True -  but this is not a good example for that.  


.....    Phil

On 2020/11/23 3:18 p.m., Phil Allison wrote:


Fair enough, under - what is it - 32VAC is unregulated...

John ;-#)#

https://www.cui.com/blog/what-is-the-difference-between-ul-listed-and-ul-re
cognized    

https://store.intellaliftparts.com/blog/ul-csa-etl-ce-abbreviations/

Per the NEC and various codes, powered item permanently installed (in the U
SA) is required to carry a UL/ETL listing.  
Technically, any mains-attached (plug-in) item sold to the public (in the U
SA) is also required to carry a UL/ETL listing. At whatever operating volta
ge.  
UL Listed items made up of sub-assemblies will typically carry UR symbols o
n those sub-assemblies. Repairs made to such items must be with UR componen
ts.  

Where this gets cute: That junk from China is sold from, and originates in  
China, is typically shipped via subsidized Chinese Post, and directly to th
e consumer - thereby avoiding the letter of regulations and codes. And then
 there are here-today-gone-tomorrow resellers that get around the code by s
imply ignoring it. Making their consumers potential victims.  

https://www.galco.com/buy/Staco-Energy/3PN1010B?source=googleshopping&utm
_source=adwords&utm_campaign=&gclid=CjwKCAiA-_L9BRBQEiwA-bm5fqpZoQpgd
sKcuyerBdXTNjaU9OAZzWB_xlXjumLQK-2_lBh92pHqrxoCQPcQAvD_BwE      This device
 carries a CSA mark.  

 https://www.ebay.com/itm/20Amp-Variac-Transformer-Variable-AC-Voltage-Regu
lator-Metered-2000VA-0-130V/124023830503   This device does not.  Note the  
difference in cost. That cost  is not only for those obscene profits on the
 part of the manufacturer, but also for proper design, proper testing, basi
c quality control, insurance and all the other unnecessary niceties avoided
 by the resellers and their suppliers. So, the bottom line is that you get  
what you pay for, with all the consequences attached thereto.  

Peter Wieck
Melrose Park, PA

On 11/24/20 9:42 AM, Peter W. wrote:



Yes, I definitely see that now.  Unfortunately, I had really hoped to be  
able to use one of these dimmers because the linear supply is going to  
take up more space, but I will not sacrifice my safety for space.  I do  
wonder about something like PC power supplies.  I have some old Dell  
ones, many times used for other purposes (those three voltage rails sure  
come in handy sometimes), but none that I can see have any UL listings.  
I've had desktop PC's run for years 24/7 without incident.  Anything  
that ever went bad was usually memory or hard drives and such.  I've had  
a lot more trouble with laptops (but still not laptop power supplies...  
motherboards!), but that's a story for another day.

https://www.cpumedics.com/dell-cpb09-000a-350w-power-supply-for-inspiron-53
0-531-vostro-400-studio-540-xps-8000-8100/?utm_source=googleshopping&utm_
medium=cse&_vsrefdom=adwords&gclid=CjwKCAiA-_L9BRBQEiwA-bm5fot0mFXYGy
4XdUEa0WJuHrkiygErfrh6ZpwTrkLXFBma4NFd4h_nExoCAlsQAvD_BwE    


https://www.cpumedics.com/dell-dk87p-240w-power-supply-with-2x-connectors-6
-pin-for-optiplex-3050-5050-7050-inspiron-3668/?utm_source=googleshopping
&utm_medium=cse&_vsrefdom=adwords&gclid=CjwKCAiA-_L9BRBQEiwA-bm5fvDBE
vqtTAvR17GLXK6yGIeqzOcCCYc4zSUHOW1AaB3vBYqqUvmrwhoCExQQAvD_BwE    

If sold in the United States:

There will not be a UL mark on a computer power supply.
There will be a UR mark on a computer powers-supply.
If there is neither, it is a knock-off.  

The power-supply is a sub-assembly.  Not the main event.  

Peter Wieck
Melrose Park, PA

On 11/24/20 10:05 AM, Peter W. wrote:


Interesting

https://www.pcscsecurity.com/wp-content/uploads/2018/04/ULversusUR_PCSC.pdf

Any external supplies I have do have the UR mark.

Well, thanks for the info.  Nice to learn something new everyday.  I  
will no longer be tempted by the lure of the cheap Chinese eBay junk.

I will say that, by all accounts, this should have been a decent dimmer  
judging by the components within but I now believe the lack of  
heatsinking was the reason for the failures.  The internal design looks  
simple enough that, before I trash it, I may try and draw out a  
schematic since most of the schematics for this junk don't exist.

On 2020/11/24 6:42 a.m., Peter W. wrote:


Thanks for taking the time to reply to such depth. I will be flagging  
this so it is easily retrieved in the future!

John :-#)#

On 11/24/20 2:36 PM, John Robertson wrote:

ul-recognized  



  

  

g&utm_source=adwords&utm_campaign=&gclid=CjwKCAiA-_L9BRBQEiwA-bm5fq
pZoQpgdsKcuyerBdXTNjaU9OAZzWB_xlXjumLQK-2_lBh92pHqrxoCQPcQAvD_BwE      

Regulator-Metered-2000VA-0-130V/124023830503    

 is not  


I appreciate this too.  Further investigation on my part might shed  
light on a cause for both of my dimmer failures:  too high of input  
voltage!  Even though advertised as 12-24VDC input, I tend not to trust  
this figure and, after looking at the components inside, I think it's  
12V, period!  Unfortunately, the driving supplies I had been using for  
this were more like for standard Ham radio, 13.8 VDC.  Perhaps this  
higher voltage could not be handled by the dimmer components.

 Chuck wrote:
============


**  You have an obvious overheating failure of one mosfet in a parallel pair.  
As mosfets heat up,  the on resistance increases by a factor of 2 or more.  
So the temp rise does as well.  
Having no heatsink at all in that device is nuts.  
Bad design, buy something else.


......  Phil  

 Chuck wrote:


** You have never posted what PSU voltage you have been using ???  

The rating of 12-24V applies to the LED array it is driving.  

You cannot use a 24V supply, connect a 12 V LED  array and set the control half way.  


.....  Phil  

On 11/24/20 5:38 PM, Phil Allison wrote:


So, it could have been either lack of heatsinking, my 1.8V overvoltage,  
or both?

Thanks for the clarification on the 12-24V.  I certainly didn't know this!

On Wednesday, November 25, 2020 at 11:35:21 AM UTC+11, Chuck wrote:

 ** Was you  8A current figure when the controller was set to full  ?  



** Yes - it is missing data in the advertising.  

A  PWM  controller is  *not* a voltage regulator,  merely a " time division" current reducer.  

If the voltage is a little high, the current will be way higher.


......  Phil