[GreenKeys] Fusing requirements for loop

[Jordan Spencer Cunningham]

I never said it melted in seconds. The loop was on for probably at least a minute-- maybe two-- before I realized what was happening (the magnets started smoking). It was after I pulled the plug that I found the component (an optoisolator) had melted. Also, I know the optoisolator worked for some period of time before it failed closed as I was relaying signals through it successfully. Probably at least 30 seconds, maybe more. Again, the other optoisolator that survived was rated for 200mA, so unless the manufacturer gave a huge margin between the official max rating and the actual max amperage it could handle, the current through the loop at the time couldn't have been too much more than 200mA.

Anyway, that is neither here nor there.

It seems like most people have found 100mA+ is still plenty safe. It's true that resistors usually fail open if they fail at all, so fusing a loop may not really be necessary unless you're an idiot like me. But I figure more safety can't hurt, especially when dealing with hard to replace components like the magnets, and especially when you're an idiot like me.

Thanks for all the input. Lots of great information here.

-
Jordan Spencer Cunningham
teletype.net


On Fri, Sep 20, 2019, at 4:07 PM, John Nagle wrote:
> On 9/20/19 12:59 PM, greenkeys-request at mailman.qth.net wrote:
> > How forgiving are selector magnets? Has anyone stress tested them?
> > Maybe even Teletype themselves?
> > 
> > I had an unfused loop burn out an optoisolator in a prototype board
> > and get up to something like 150 - 200mA. My 15's magnets started
> > smoking. But since it all happened so quickly and I shut off power
> > when I realized what was happening, I may not have measured the
> > current properly, but I thought I glimpsed something in that range.
> > Also, the melted component in question was rated for 150mA. The other
> > component on the board that did NOT burn out was rated for 200mA, I
> > believe. So the current had to be somewhere in that range, and I know
> > the magnets were getting dangerously hot.
> 
> If you melted a 150mA component in seconds, you were way over 150mA.
> Probably several times that.
> 
> My early interface board used a largish open-frame 120VDC supply
> I bought in a surplus store. I used a 125mA fast-blow fuse on that.
> Never blew one.
> 
> My USB-port powered interface board has no fuses, but some protection
> circuitry.
> 
> https://raw.githubusercontent.com/John-Nagle/ttyloopdriver/master/board/images/schematic.png
> 
> That unit has a custom switcher which charges up 2uF to 120V to provide
> the initial current needed to push though the 5 Henries or so of
> inductance in the selector magnet. That supply takes a whole SPACE
> time to charge up, and can't deliver enough current continuously to
> cause overheating.
> 
> The sustain supply, which keeps current flowing after the pull-in
> pulse is from U8, a 5VDC to 15VDC DC-DC converter. That's only rated
> for 66mA at 15 volts, so it can't produce enough output to cause
> overheating, either.
> 
> At the output end, Q2 and R5 form a 60mA current limiter.
> That can survive a short circuit indefinitely; Q2 will heat up
> but it stays well within its thermal limits. I checked with an
> IR thermometer. That's the main short circuit protection,
> more to protect the electronics rather than the Teletype.
> 
> At the input end, there's a AP2553W6 USB power interface part.
> That limits current drawn from the USB side to 400mA, and
> that's mostly to prevent the protection circuitry in laptops
> from shutting down the power. (If you draw too much power
> from a USB port, even for a millisecond, it turns off
> power to prevent crashing the laptop. Which is why you
> need a power interface part with current limiting on
> anything that has a filter cap which needs to charge up
> at power up.)
> 
> John Nagle
> 
> 
> ______________________________________________________________
> GreenKeys mailing list
> Home: http://mailman.qth.net/mailman/listinfo/greenkeys
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:GreenKeys at mailman.qth.net
> 
> 2002-to-present greenkeys archive: http://mailman.qth.net/pipermail/greenkeys/
> 1998-to-2001 greenkeys archive: http://mailman.qth.net/archive/greenkeys/greenkeys.html
> Randy Guttery's 2001-to-2009 GreenKeys Search Tool: http://comcents.com/tty/greenkeyssearch.html
> 
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to js at teletype.net
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qth.net/pipermail/greenkeys/attachments/20190923/caf50efd/attachment.html>