On Friday 07 February 2025 04:05:02 pm wn4isx via groups.io wrote:
On Fri, Feb 7, 2025 at 02:02 PM, Roy J. Tellason, Sr. wrote:
The overvoltage circuit at ( ) are all effective but the one with the PNP transistor and SCR are the most stable as far as trigger point.
Okay...
If I where going to use a TRIAC, I'd use it in SCR mode with the PNP transistor.
In addition to having some triacs on hand, the thought also occurs to me to use a complementary cross-coupled pair of transistors in place of that SCR. Collector of each one is tied to the base of the other one.
You can vary the value of the capacitor C1 to achieve different firing delays. There are systems that can tolerate a few milliseconds of overvoltage.
I'm not sure how you'd tell what something would tolerate. "Absolute maximum" ratings for say CMOS parts is absolute maximum...
Reduce the value to speed up the delay, remove it for super fast but be aware the circuit might be prone to tripping on exterior events like nearby lightning.
They mention using a small inductor in that web page. I'm thinking this might be a good place to use some of those ferrite beads I've accumulated.
I use 50nF and I'm not bothered by falses but it still trips if I emulate a regulator failure.
How do you do that?
Call me paranoid but when I built a linear power supply for a Perseus SDR (~1K in 2000) I used two OVP crowbars in series with two SCRs at the output of the filter capacitor and two SCRs, one each, at the output of the OVP.
I'm reminded of a service call I did once. The company made paper filters for all sorts of commercial applications. The output of the paper making process was rolls of paper, which were stacking up and they were running out of room to put them in. The machine I was there to work on would take a roll of paper and punch big circles out of it. The electronics was a couple of boards full of 4000-series CMOS parts, and the power supply was driven by a variac! One sensor was cabled in the same conduit as a 240V motor drive cable, I wanted to remove the wire from inside of that conduit but was overruled. There were two sets of boards but they were somewhat different revisions so you had to swap them both, if you were sure that you had a good set, which they weren't. It was me, another contractor from a different company, and the "fixer" who was the guy working there who was supposed to keep stuff running. Then they flew out a guy from the factory that mde this thing, which is when we found out that the boards were different revisions. I did sketch out a crowbar circuit to the fixer, told him that they needed something like that in there, I don't know if they ended up building one or not. They wanted to know if I stocked piles of 4000-series CMOS parts, I didn't at that time,
I think bad power supply design, no overvoltage protection, no indication of the boards being different contributed to their situation. I was out there for a couple of days, dunno how long it took them to get it finally fixed...
--
Member of the toughest, meanest, deadliest, most unrelenting -- and
ablest -- form of life in this section of space, ?a critter that can
be killed but can't be tamed. ?--Robert A. Heinlein, "The Puppet Masters"
-
Information is more dangerous than cannon to a society ruled by lies. --James
M Dakin
