[bars] When is a fuse not a fuse?

Mon Aug 22 11:06:05 EDT 2022

Just 7 years of normal thermal cycles with stranded wire (my electrician does not like stranded wire for these applications, but he didn't install it).  Finding a 6-foot length of 6ga wire was hard enough that I'd rather check the screws every year than try to rewire it (and then still check the screws every year).  8*)

The inverter's rated for 10KW at 240V, so 42 amps(ish), and the circuit has a 60A breaker (same size as the main for this building).  Grid-tie inverter, so it's made for this application, and seems to work well otherwise.  

No noticable HF interference, but then 40M isn't very good in the daytime, so I might not notice.  The one-inverter-per-panel systems apparently generate a lot more RFI in aggregate, but the island array (which may be that style) is 1/2 mile away, so fongers crossed...

The left-hand wires are the DC from the array (400V nominal, 600V peak, 25-ish amps), and the right-hand wires are the grid connection.  All of the screws were not-tight, though they are now!

73, Willie N1JBJ

> On Aug 22, 2022, at 10:17 AM, Geoffrey Feldman <geoffreyf at comcast.net <mailto:geoffreyf at comcast.net>> wrote:
> 
> Screws loosen due to overheating.  Something else is going on leading to this loosening.   People fear shorts in electrical equipment but the greater danger is arcing from loose connections.
>  
> What is the maximum current rating for this?  Are you exceeding that. Also inverters that don’t produce a good approximation of a sine wave or are out of phase with the mains power will heat up.  Something else is wrong.  Find the cause behind the symptom.
>  
> Interestingly you may find some interference on your radio receivers that gets worse when the inverter is functioning.  
>  
> Hard to say from the photo but one of those black wires don’t look too healthy either.   Where do these wires go?
>  
> Geoff
> W1GCF
>  
> From: William Smith [mailto:w_smith at compusmiths.com <mailto:w_smith at compusmiths.com>] 
> Sent: Monday, August 22, 2022 10:03 AM
> To: Geoffrey Feldman
> Cc: Tim KC1RET; BARS Reflector
> Subject: Re: [bars] When is a fuse not a fuse?
>  
> Indeed, I had a solar inverter blow up and short it's output, clearly a semiconductor failure, as my DVM didn't measure any significant conductivity between either leg and ground, but turning the power on immediately tripped the 60A 240V main breaker for that building.
>  
> Went back and forth with the vendor, finally opened up the AC power connection point:
>  
> <image001.jpg>
>  
> See the red wire third from the right?  Yes, that red wire:
>  
> <image002.jpg>
>  
> Turned out the screws got loose over time, and the insulation got carbonized by the heat till it shorted out.  Swapped the wire, tightened the screws, and it's fine.  Also put a sticker on the box to remind me to (re)tighten the screws yearly.
>  
> When is a short not a short?
>  
> 73, Willie N1JBJ
>  
>> On Aug 22, 2022, at 9:50 AM, Geoffrey Feldman <geoffreyf at comcast.net <mailto:geoffreyf at comcast.net>> wrote:
>>  
>> It would be interesting to know what “The fuse” really is and what purpose it is there to serve.  Semiconductor junctions can change when stressed.
>>  
>> This brings up something which I feel is really important in electronics.   Terms such as “Resistor”, “Capacitor”, “Transistor”, “Diode” , “Inductor” are really ideals of what the thing really is.   Capacitors have inductance. Inductors have capacitance.  Transistors and diodes have capacitance.  Resistors have inductive properties. Conductors together on a board have all of the above.  We need the ideals to simplify our thinking but we need to know – there is more to it than that.  By the way, not all fuses are alike and none of the above die a natural death.
>>  
>> A favorite experiment is that LED’s actually can produce current when exposed to light.  This actually is important in developing LED lighting because light reflecting back into the LED can increase current and reduce efficiency.   The other fun thing is that LED’s are sensitive to light in a near but not exact wave length to what they produce.    A fun Arduino type project is to pulse an LED just a the margin of producing light.  This builds a charge in the junction.  The Arduino can measure this charge. The charge dissipates quicker if the LED is exposed to light near the wave length it emits.   So where are the ideals now?
>>  
>> This type of clip and measure process is one that I have used myself. Though, It can produce a poor customer response when the thing is “Fixed”.  A more patient approach of observing whether voltages make sense on a schematic is another option.
>>  
>> W1GCF
>> Geoff
>>  
>> From: bars [mailto:bars-bounces at w1hh.org <mailto:bars-bounces at w1hh.org>] On Behalf Of Tim KC1RET
>> Sent: Monday, August 22, 2022 9:35 AM
>> To: BARS Reflector
>> Subject: [bars] When is a fuse not a fuse?
>>  
>> The saga of my 'dead' TS-2000 continues.
>> 
>> So after hours of deciphering schematics and finally performing a radio autopsy on my kitchen counter I was struck with this conundrum... How could the supply voltage read 13.8v at the connector, but only 5.2 where the wire connected to the main board?
>> 
>> At first I suspected the filter caps. They are soldered directly to the back of the connector inside the radio, dead-bug style. I cut they leads, one-by-one and they were all good & no change to the voltage reading. In a fit of desperation, I checked continuity between the board and the back of the PS; + was good, GND, not-so-much. Huh?? It was measuring 2.5k ohm! I disconnected the PS lead from the radio, 2.5k between the PS lead connector and PS terminal. Ah HA! (he says - well, I really said something else quite a bit more expressive but I'll leave the specifics to your imagination) I pulled the inline fuse out - Yup, it read not open, not short, but 2.5k. I replace the fuse and the rig fired right up!
>> 
>> So, when is a fuse not a fuse? When it is a resistor!
>> 
>> That is a new experience for me -- I'm def. NOT going to forget it.
>> 
>> 73,
>> 
>> T.
>> 
> 
>  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.w1hh.org/pipermail/bars_w1hh.org/attachments/20220822/8f0023c8/attachment.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: IMG_9888.jpeg
Type: image/jpeg
Size: 80023 bytes
Desc: not available
URL: <http://mail.w1hh.org/pipermail/bars_w1hh.org/attachments/20220822/8f0023c8/attachment.jpeg>