Electronic Hipp Errors ?

I have been following Carl Wilson's article in ME on an "Electronic Hipp Toggle Circuit" with the idea of using it with some modifications. I plan on using Hall effect devices vice opto interrupters. This will require some changes to the logic as the output of the Hall devices is high in the unenergized state whereas the opto ones are low but I haven't decided how I will implement it yet.
In going through the schematic, the block diagram and description of the logic flow, unless I am misinterpreting something, it seems to me there is are some errors in the schematic as drawn as it doesn't agree with the block diagram and the logic description.

The way the schematic is drawn the outputs of the three flip flops appear to be backwards. The output of the "MIN" F/F is shown as the "Q" output but the block dia. and description both show it as the "NOT Q" output. Same for the ENA and ON F/Fs both shown as "NOT Q" where a the block dia. and description show both as the "Q" output. (The ON F/F does show the output labeled as "Q" but is in the position of the "NOT Q" output )

The description and the block diagram appear to agree.

John

Edited By John Purdy on 31/10/2022 02:50:23

Edited By John Purdy on 31/10/2022 02:54:39

Please disregard the above post! The schematic is right, I wasn't thinking straight!

If a moderator sees fit you can delete this and the above post. Thanks.

John

Irrespective of whether the thread is to be disregarded, are not “Logic Gate” chips designed for this sort of thing?

NOR and AND gates, etc?

John, one good reason to use optos is that they are MUCH more precise than Hall effect sensors. Optos can sense to sub micron precision but HEDs sense a magnetic field that extends into space. I've given some results on here for optos and can also provide some for HED when I'm on my proper computer.

https://www.model-engineer.co.uk/forums/postings.asp?th=171996

Edited By John Haine on 31/10/2022 07:33:58

It's a lot easier using a processor. My clock uses 2 optos and a Pic, but it could easily be rewritten for Arduino. With a bit more thought it could get down to one but that would have the processor doing timing, which I tried to avoid, I want the pendulum doing timing

Edited By duncan webster on 31/10/2022 13:37:06

John, the HEDs I'm planning on using are unipolar devices with a built in trigger circuit, so the OC output transistor should switch cleanly. (If not satisfactory I have a drawer full of optos to try ).

Duncan, the reason I'm going to try the discrete route is I have a box full of 74XX, 40XX and 45XX chips so I though I might as well use them. If I get fed up with that route I can always go the micro route as I have a spare Arduino to try.
John

Here is a Hall effect sensor that I'm using to update an electrically-maintained clock, replacing a platinum contact. The actual sensor is the 3-legged black box in the foreground on the top. Somewhere I've noted the type, it's made by Allegro and is a "continuous time switch" driven with 5V with an output that goes low and lights up when a field in excess of a threshold is sensed. The product came on a "Linker" compatible PCB for Arduino from Maplins (still available on eBay). The board is mounted on a Corian pillar with a steel base that has a neodymium magnet embedded in it holding it on to a steel mounting plate. The magnet is embedded in steel and there isn't enough stray field to give any perceptible effect on the sensor. There is also a 2x3mm (d x l) neo magnet in the "finial" sticking out of the base of the bob. With a 5mm gap between the magnet and the sensor, the latter is triggered when the magnet is ~1.2mm from the vertical axis and releases when it's ~4mm away. So of the order of 3mm hysteresis, and these numbers are very dependent on the height since the field of the bar magnet varies as the inverse-cube of distance. The hysteresis is probably electronic and a discrete analogue HED with comparator might improve on that.

For this clock these numbers are acceptable since it will only be used to activate an electromagnet that "arms" a gravity escapement - so the precise impulse timing is determined by the mechanism geometry. However I suspect that if an HED was used for the primary impulse timing the precise phase and length of the impulse would not be very precise. For that reason I prefer opto interrupters, especially a type made by Sharp that has a built-in schmitt trigger.