SLC 5/04 Performance

Hi ... I'm new here so please be gentle.

Could anyone estimate the maximum pulse frequency obtainable from an SLC 5/04 (541) CPU and OB16 transistor output card. The amount of code other than the pulse generator part can be considered negligible.

Regards
Rob

I'm not sure what the fastest scan time a slc can achieve. I've got a test rack set up in my office. I'll reply tomorrow to let you know what to expect for minimum scan time, which I'm sure you know would limit the the max frequency your plc would update an input on your card.

Remember that it takes about a millisecond to turn off a 1746-OB16 output circuit, so your ON duration minimum will be about 2 milliseconds. You can set a Selectable Timed Interrupt (STI) for as low as 1 millisecond on an SLC-5/04... so I imagine you could generate a 4 millisecond square wave with a SLC-5/04 and an OB16 module. I'm sure you'll get some more empirical evidence from subsequent postings.

What do you need to do with the pulse output ? There might be a better controller for your application. The MicroLogix 1200 and 1500, for example, have a special PTO pulse train output function.

You can get faster switching times if you use the TTL (1746-OG16)module instead of the DC output module.

It is possible to turn an output on and off in an STI scan as opposed to the classical oscillator that toggles states every scan. We do this all the time in our industrial control applications. In the first rung of the STI file, latch an output and then perform an immediate output. Next perform some additional ladder code which can be bogus code (see below), and then in the last rung of the STI file unlatch the output and perform an immediate output instruction again. You will have to experiment with the STI times to determine how fast you can run. You will need some code (even if it is bogus code) to give the output time to turn on before you tell it to turn off. A good choice for bogus code is to insert one or more "X to the power of Y" instructions on a couple of floating registers to intentionally cause a slight delay between switching the output on and switching it off again. If you observe the pulse train with a scope, you will see a pulse train with a period equal to your STI rate, and with a duty cycle somewhat (not exactly) proportional to the amount of code and overhead in your STI file.

For best results, set the interrupt latency bit S:33/8 (under the STI setup tab) to a value of '1' prior to downloading your program. Make sure you have some minimum load connected to your output i.e. a 10K resistor works well. Otherwise you will observe strange 'turn-off' times do to the high impedance of your scope leads.