74HC165 12 volt current limit

Go To Last Post
12 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I am building some new input boards for a pipe organ keyer.

In testing the existing resistor buffer to the 165 I am finding I do not understand why the old analog engineers that designed these interfaces back in the 1970s apply 12 volts (actually 13.5V from the unloaded supply) to the input pin, rather than using the resistors as a divider.

I looked at several designs and this seems to be the way it is always done.


 +12 Volts when key depressed, otherwise open
 |
 |    100K
 *----/\/\/\-----> to 'HC165 input
 |
 |
 >
 <
 10K
 >
 <
 |
 |
 - com or negative return

When I measure the voltage at the input to the shift register, it looks close to 12V. So does the 100K current limit this? How is it possible to put more than 5 volts on the input pin?

I think in the 1970s they were using straight 74C*** chips as some seem to use + and - 5 volts. But the more recent boards that use this method defiantly use the HC variants.

What is the advantage of doing it this way rather than using the resistors as a divider like in some of the examples posted in the archives here?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

4000 logic and then 74Cxx run from 3V to 15V (16V max) so 12V was being kind to the chip. :-)

You only get decent switching speeds at higher voltages. Not very fast at 5V.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

js wrote:
...and then 74Cxx run from 3V to 15V (16V max)
74HCxx - never, only CD4xxx.

Warning: Grumpy Old Chuff. Reading this post may severely damage your mental health.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I NEVER said HC only 74Cxx. :-)

edit in the circuit shown they are simply using the internal diodes to clamp the voltage at VCC. With 100K in series that's very little current.

John Samperi

Ampertronics Pty. Ltd.

www.ampertronics.com.au

* Electronic Design * Custom Products * Contract Assembly

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

With an organ key sensor, speed is not much of an issue. 1ms response speed would be way-fast.

IF the logic is powered from 5V, then I agree with John. The 100K resistor in combination with the high-side input clamp diode effectively limits the input voltage. If the logic is power is 12V, then the resistor probably protects the logic input from ESD or something like that.

The 10K resistor provides a little switching current to keep the switch contacts good. Mechanical switches are not reliable with zero current, even gold plated contacts. The 1ma to 1.5ma switch current is enough to keep the oxide layer to a minimum on the contact and results in more consistent switching.

As an aside, this same exact circuit can be used with a 5V GPIO pin, just fine.

Jim

 

Until Black Lives Matter, we do not have "All Lives Matter"!

 

 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

js wrote:
4000 logic and then 74Cxx run from 3V to 15V (16V max) so 12V was being kind to the chip. :-)

You only get decent switching speeds at higher voltages. Not very fast at 5V.

The keys are all open contacts (usually phosphor bronze) Basically a shorting bar closes the circuit. That part goes back to the late 1800s. In those systems it was all mechanical relays, which moved banks of contacts.

The old organ from the 1970s uses 74C** series. These were really sensitive to static. Especially with the open contacts. I figured the higher voltage was to make sure that the signal was read.

A lot of the systems use what is called positive return, where the frame is actually the plus side of the system. That part make sense when switching the coils to the active state. Interesting that much of this was based on early telephone switching systems.

From my reading in these forums, I thought clamping diodes might have something to do with it. Was not sure if that was an AVR input pin feature or the way that the modern (H)cmos inputs work.

Thanks for that clarification, still would like to know what the advantage is of doing it this way.

The pipe organs are pretty noisy environments. With a lot of vibration inside, especially when the trems and regulators are spilling wind. Then the pipes go from 16Htz to above what the ear can hear.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
, still would like to know what the advantage is of doing it this way.
Quote:
The 10K resistor provides a little switching current to keep the switch contacts good. Mechanical switches are not reliable with zero current,

The gate alone doesn't draw enough current to keep the contacts clean.

JC

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

jporter wrote:
A lot of the systems use what is called positive return, where the frame is actually the plus side of the system. That part make sense when switching the coils to the active state. Interesting that much of this was based on early telephone switching systems.

That's quite logical, since the old electromechanical telephone systems ran at minus 50 volts, positive ground. Anyone used to working with telephone technology would wire it that way without thinking about it.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:

... the old electromechanical telephone systems ...

Perhaps it is also why we might use energy-meter chips for current/voltage work in our AVR apps, or audio codecs--they are designed for a mass-market and we can get a few of these very sophisticated devices for our own work at a very reasonable cost.

Telephone switching gear was probably readily available in a variety of sizes, as not only the telephone company would have it in switching stations but businesses of various sizes would have a setup as well.

The sparkies will correct me, but higher voltages+higher currents ==> higher noise immunity. (right?)

You can put lipstick on a pig, but it is still a pig.

I've never met a pig I didn't like, as long as you have some salt and pepper.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

theusch wrote:

Perhaps it is also why we might use energy-meter chips for current/voltage work in our AVR apps, or audio codecs--they are designed for a mass-market and we can get a few of these very sophisticated devices for our own work at a very reasonable cost.

It wasn't just that telephone tech was available - it was the only show in town. Back in the day, telephone engineers were the alpha geeks. If you needed someone to design an electric organ keyboard, they were the only people with the training and knowledge. That probably lasted until about 1960, when solid state electronics appeared. It was telephone engineers that designed and built the first code cracking computers in WW2, for example, and the 1948 WITCH computer used banks of Strowger 2-motion selectors, the tall cases in the center of the picture. The tubes are mostly decatrons.

Attachment(s): 

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

What are they doing, reading paper tape? By hand?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I think they're just posing for a photograph. But back in the day, some people could read it by eye. That will be 5-hole Baudot (Telex) tape in the picture and it's probably all numbers for this machine, so it wouldn't be that difficult.

here's the original text to go with that picture:

Quote:
The picture shows our earliest computer called the WITCH which stood for Wolverhampton Instrument for Teaching Computing from Harwell. It was built at AERE Harwell in 1948 and was won by the then Wolverhampton and South Staffordshire Technical College in a national competition in 1957. The photograph above was taken in 1961.

The technology consisted of mixture of 1940's vintage telephone exchange technology (GPO 3000 type relays) and early nuclear instrumentation (Dekatron tubes).

A Dekatron tube was a device rather similar to a neon indicator light except that it had 10 anodes rather than one. By carefully timed and shaped steering pulses on intermediate electrodes the active glow could be moved from one anode to another. Dekatrons were used for event counting in early nuclear instrumentation before the advent of solid state electronics. In the WITCH the Dekatrons were used for storing digits of numbers. A row of 10 dekatrons could store a single number.

The WITCH was a very slow computer by modern standards. It took it

* 2 seconds to add or subtract 2 numbers
* 5 seconds to multiply two numbers
* 15 seconds to divide two numbers. (Division by zero took rather longer.)

The fully developed configuration at Wolverhampton had 90 memory locations. Programs were read from a paper tape reader in an adjacent room.

There were actually 6 electromechanical paper tape readers, the current one being selectable under program control. Whenever the WITCH had finished executing an instruction it read the next instruction from the current paper tape reader. Program loops were constructed with the aid of a pot of glue.

One unfortunate problem with paper tape loops was the tendency of the mechanical readers to poke extra holes in the paper tape after several passes. If your job was sufficiently important you could use special linen tape which was more resistant to this effect.

It was possible to determine the contents of any memory location by simply exmaining the relevant row of Dekatrons. The storage locations were in the two racks visible on the left hand side of the picture. The central rack of electronics contained the arithmetic and control units. It was possible to watch the multiplication of two numbers by seeing the partial products building up in the accumulator (another row of Dekatrons).

The WITCH was used for many years for introductory and schools' courses. In its later years it became increasingly unreliable and spares become difficult to obtain. Eventually, in the mid 1970s, it was retired to the Birmngham Science Museum, where it is still on display.