interruption of power

[dano]

I would like to build a circuit in which I could interrupt the power to a PC for an adjustable length of time. Voltage is 120 volts AC and the length of time would be 500 ms down to below 1 ms if possible.
Anyone have any ideas?

Thanks
Dan

 

[Nigel Goodwin]

I would like to build a circuit in which I could interrupt the power to a PC for an adjustable length of time. Voltage is 120 volts AC and the length of time would be 500 ms down to below 1 ms if possible.
Anyone have any ideas?

It won't come as any surprise to most people when I suggest a PIC to do it, but a 555 could be easily used as well - but a PIC would be far more accurate.

But timings not going to be the problem, getting very short interuptions is the problem.

I'm assuming your mains is 60Hz?, so a zero crossing point occurs every 8.33mS (every 10mS at 50Hz) - if you use a triac it would be difficult to make a break shorter than that. What you could do is have a small interruption immediately following a zero crossing point, but a 1mS break would hardly be noticeable. But this is another area where a PIC would 'wipe the floor' with a 555.

Using a relay would probably be worse?.

I'm presuming this is intended to test the susceptability of a PC to mains interruptions?. I suspect missing a complete half cycle (8.3mS) would have very little effect on the power supply.

 

[dano]

Wow, Nigel you're quick. Thanks for the reply.

Yes this is to test the susceptability of the PC to mains interuptions. The exact application is as follows: I have a CNC milling machine which is dropping out of the operating software and reverting back to DOS. I would like to test the possibility that it is caused by an interruption of power, thus the reason for the question. The problem occurs at random, it has gone as long as several weeks without failing.

How long of an interruption will a PC tolerate?

Dan

 

[ChrisP]

Dan -- it would be far easier, if we are simply talking about a PC, to power the PC via a UPS... and I would recommend a double-conversion online UPS for this application.

 

[gerty]

A UPS will also filter the spikes that might be generated from the machine, (contactors,relays,drives)
We use UPS on a lot of our stuff here with good results..

 

[dano]

Let me first say that this forum is great. I look forward every day to reading everything which is posted. Thanks.

I've been looking into the UPS option.

I'm the kind of guy that likes to find the root of a problem, thats the reason for wanting to find a circuit to interrupt the power. I don't like throwing parts at a machine without really understanding what's going on.
I may have to install a UPS just to keep the customer happy, the problem is that the machine is under warranty, so the company I work for eats the cost unless I can prove it is the incoming power.

The customer who has the problem has two machines, both have failed in similar ways. I have many other machines at other shops which have not failed in this way. That is what makes me think it may be a power issue>

Dan

 

[ChrisP]

...So monitor and log the incoming power at the device to get your proof.

Either way, the customer would be well advised to use a UPS or line conditioner to protect his equipment investment. If line variations have already "killed" two units, doesn't it stand to reason that deliberately introducing line problems -- even for testing -- could damage the equipment?

 

[dano]

ChrisP. when I say the machines fail what I mean is that the computer must be rebooted to get back to working again. The problem is the time lost (even if it is only a few minutes) during production. For test purposes interrupting the incoming power would cause no harm to the equipment .
The electric utility is monitoring the incoming power as I type this. I hope we have a failure during the test. If we can't resolve the issue in this way then I'll take your advice on the UPS.

The other sticky issue is we are not the manufacturer of the machines, only a dealer.

Another reason for wanting to make this type of switch is because I'm curious.


Dan

 

[ChrisP]

Don't get me wrong -- I fully understand the curiosity -- I'm very much the same way myself!

A couple of things to consider here:

• You said that the power utility is monitoring the power. All well and good... but are they monitoring it incoming at the facility (more likely), or are they monitoring it at the affected equipment (less likely). This makes a huge difference if the problem has a local cause, e.g. a heavy spike or sag caused by some other installed equipment.
• The answer to your question regarding the vulnerability of PC's to line problems really is almost unanswerable. This is because of the vast differences in PC power supplies. Different supplies can and do respond differently to line problems. As an example, of the five PC's operating in my shop, two tend to reset with very short duration power loss, while the remaining three don't. it is precisely for this reason that line protection (read:UPS) becomes very important in establishing any kind of uptime reliability.
• There are of course other factors that could come into play with your problem. Some notable ones are power supply temp, system/CPU temp, and operating system in use.

 

[dano]

Chris, the power is being monitored at the machine.

thanks for the replies.