# How the heck does this work?

Joined Sep 26, 2006
12
Can anybody tell me how this can possibly work? I need a super simple constant current supply from a rectified AC line. I came across this circuit in patent number 7,508,240. (http://www.google.com/patents/US7508240) I have no need for a zero crossing detector, as in the patent, but I do need a constant current source. How can simply putting a zener in parallel with the LED of an optocoupler work at this voltage level? I would figure the LED and zener would blow up if hooked to 1/2 wave pulsed 240VDC input. If this does work, is there any way to get some idea of what kind of zener and at what voltage? Is there any way to make this circuit provide about 15ma from the rectified line voltage? Thanks everyone.

#### strantor

Joined Oct 3, 2010
5,508
You know, just because it's patented doesn't mean it works.

#### GopherT

Joined Nov 23, 2012
8,012
Also, not all drawings are there to describe the invention, they could be representing a common problem that will be solved with the invention. It is dangerous to look only at the drawings when their is plenty of text available. Sometimes a picture is worth a 1000 words, but it doesn't help at all if the context of the story is not available.

Be careful with this circuit, it is essentially a transformer-less power supply.

#### joeyd999

Joined Jun 6, 2011
4,477
As the inventor of six patents myself, I am still amazed at some of the things that can be patented! A zero-cross detector? I'm sure this particular patent could be invalidated through a relatively quick prior-art search.

As disclosed, I don't think the zener current source functions as the inventor thinks. Regardless, the first independent claim does not require a "zener current source", just a current source of any technology powered by rectified AC:

1. A zero-crossing detector, said detector comprising:
a pair of input terminals, said terminals being adapted to receive an AC input signal;
a rectifier, said rectifier rectifying said AC input signal;
a constant current source, said constant current source being powered by said rectified AC input signal; and
an optoelectric coupler having a coupler input and a coupler output, said coupler input being driven by said constant current source and said coupler output providing a zero-crossing signal, said zero-crossing signal being galvanicly isolated from said AC input signal.
Therefore, the validity of the patent does not depend on his zener current source as disclosed to be functional.

Interestingly, the patent is assigned to Keithly Instruments. They probably paid their engineer to invent the "technology" and patent it, and it is probably useful for them in some form. But I wouldn't want to bet my business on a patent like this holding up in court.

Remember, though, that in the end, a patent is only worth as much as you are willing to pay to defend it.

#### wayneh

Joined Sep 9, 2010
17,061
I need a super simple constant current supply from a rectified AC line.
How much current i what voltage range? Constant current is pretty simple. I can say that because even I have built one.
What do you mean by "super simple"?

#### joeyd999

Joined Jun 6, 2011
4,477
BTW, I had "invented", years ago, a low-noise, high accuracy 300V 30mA current source using an LM317HV. It included an optocoupler to tell you when it was regulating properly. I won't tell you how to do it, since you intend to run it off the mains -- and that can be dangerous.

Joined Sep 26, 2006
12
I did not think it would work either. I'm trying to make a line powered strobe simmer circuit. These circuits are commonly line powered and I've been doing this for the better part of 30 years. I know these kinds of things can be dangerous. There really isn't anything much more dangerous than a 3000uf bank of caps charged up to 800V. I haven't killed myself or anybody else....yet. I need to make a simple circuit with low parts count. I'm limited on space. It needs to supply at least 15ma at 320VDC to keep a strobe tube warmed up between pulses.

#### joeyd999

Joined Jun 6, 2011
4,477
...I need to make a simple circuit with low parts count. I'm limited on space. It needs to supply at least 15ma at 320VDC to keep a strobe tube warmed up between pulses.
I will dig in my paperwork tonight to see if I can find the circuit...on the condition that you recognize that it was designed to work on the far side of an isolation transformer.

#### joeyd999

Joined Jun 6, 2011
4,477
I went through all my old stuff last night. Unfortunately, I designed that current source back around 1994 or so, and moved twice since then. I cannot find a copy.

More frustrating, though, is that while I remember precisely how it worked, I cannot remember how to implement it! It makes me realize how much I once knew that I have now since forgot.

#### Ron H

Joined Apr 14, 2005
7,014
Also, not all drawings are there to describe the invention, they could be representing a common problem that will be solved with the invention. It is dangerous to look only at the drawings when their is plenty of text available. Sometimes a picture is worth a 1000 words, but it doesn't help at all if the context of the story is not available.

Be careful with this circuit, it is essentially a transformer-less power supply.
Here is the text describing the operation of Fig. 3.
Referring to FIG. 3, an example of an implementation of the current source 126 is formed from a Zener diode 126′ in parallel with the diode 122A. As long as the voltage across the Zener diode 126′ is not below the Zener voltage, the current through the diode 122A will be constant.
Hogwash.
By adding a couple of resistors, you could make a passable current source.

#### Attachments

• 10 KB Views: 10

#### joeyd999

Joined Jun 6, 2011
4,477
OK, so this has been bugging me all day. I dug deep into my memory, and I think I got most of the design re-figured out:

Basically, Q1 & Q2 maintain about 10V across the LM317, while R4 sets the current (~1.25V/R4). All the excess voltage is dropped across Q1, so it must be able to sustain the max voltage and power (4.5W worst case).

You may need to tweek some component values...I just came up with these in my head real fast. Oh, and IIRC, there needs to be a small cap somewhere for stability, though I don't remember where or what value. Perhaps across R3, maybe.

#### Attachments

• 63.9 KB Views: 31

#### joeyd999

Joined Jun 6, 2011
4,477
Oh, and by the way, the way to test this is to tie a milliamp meter from the output to ground and gradually increase the voltage. Confirm that the current remains around 15 mA and the voltage across the LM317 does not exceed 40V (it should stay around 10!). Also, make sure the junction temperature of Q1 stays below 150C for your highest expected ambient temperature.

And, in hindsight, R1 may dissipate up to 3W worst case.

#### LDC3

Joined Apr 27, 2013
924
Those are low voltage solutions.
That's true, but the OP never stated what voltage he wanted, other than saying that the diode in the original schematic might blow up with a 240V source.

#### joeyd999

Joined Jun 6, 2011
4,477
That's true, but the OP never stated what voltage he wanted, other than saying that the diode in the original schematic might blow up with a 240V source.
Sure he did:

...It needs to supply at least 15ma at 320VDC to keep a strobe tube warmed up between pulses.