Can I buffer the effects of inrush?

Jay,

I have the same problem with a laser printer. I eventually put it on a separate circuit so it wouldn't interfere with other stuff.

The UPS still may be a good solution, but keep in mind that there are (at least) two types of UPSi.

The cheaper ones feed the output from the input until an undervoltage is detected. Then, their outputs switch over to the internal supply. This type won't help your problem.

There is another type and I think it's referred to as an "online" UPS. It always feeds its outputs from the internal supply and uses the input voltage only to charge its batteries. This may provide the buffering you need, but you'd have to try it to see if the results are acceptable.

Bob

Probably the simplest off the shelf solution: a desk lamp plugged into a small UPS. The other stuff will still dim each time the laser printer's fuser heats, but your lamp will work from the UPS.

Ed

desk lamp.

There is a device called an Inrush Current Limiter. It is inserted in series with the input line. The common use is to limit surge current to power supply capacitors. The device starts at a high resistance; as an example 5 ohms, then as it heats up the resistance drops down to .2 ohms. They come in many different values. I don't know if it will work for your application, but they are very cheap and worth a try. See;

Check out the FAQ. Mike

See the last item on this list of parts. # SL32 10015

Those are NTC's and react slowly... His line or service to his complex is under rated for the load. Putting in soft start circuits would most likely cause the printer to malfunction.

--
"I\'d rather have a bottle in front of me than a frontal lobotomy"
http://webpages.charter.net/jamie_5

Check the info on the website, a quote from one page,

" Typical resistance ranges are from 0.25 ohm to 220 ohm, depending on the amount of protection desired. As current flows through to the Surge Limiter, the resistance drops in milliseconds to as low as 0.01 ohm, which functionally removes it from the circuit."

I'm sure this depends on the particular part, But I think these devices react quickly. Mike

Even if they worked quickly, wouldn't it still cause problems with the printer? I'm assuming that, whatever the printer's doing - and let's call it "heating the fuser" just so we have something less abstract to talk about - it actually needs the current it's drawing. So limiting its current draw would just cause it to not be able to heat the fuser, right? Kinda like preventing a diabetic from finishing the orange juice by locking the fridge. Yeah, it preserves the OJ, but...

All of the "inrush protectors" I've seen, including this one, are designed to protect a given device *from* inrushing current. In my case, though, the printer (according to Xerox) is operating correctly - when it's heating up, it draws a higher-than-average current, because it needs it. (Kinda like "locked rotor amps" on a motor.) So I don't want to protect it from itself; I just want everything else to not notice it.

I'm assuming what I need is a clever device that detects the large current draw, goes back in time, and builds up enough reserve power to amortize that current draw over time, so the upstream circuit doesn't see it. But, since we can't do that, it'd have to be something that's online all the time, drawing slightly more current than it really needs, and storing it up.

In essence, it's an online UPS, but from what I read about UPS's, they aren't good at this particular task for whatever reason (and here's where it would help if I understood electricity).

--
Jay Levitt                |
Boston, MA                | My character doesn\'t like it when they
Faster: jay at jay dot fm | cry or shout or hit.
http://www.jay.fm         | - Kristoffer

You need a better service for the printer.

--
"I\'d rather have a bottle in front of me than a frontal lobotomy"
http://webpages.charter.net/jamie_5

Ok, sounds like the lights dim and stay dimmed until the printer shuts down, not just dimmed during printer startup. Find the outlet closest to the circuit breaker and plug your printer into that outlet. See if that helps. Mike

Yeah, I think you're right. I did a little more searching - funny how you can find different things on different days - and I found an article:

It may be a little dated, but the upshot is:

• An online UPS *can* buffer the effects of inrush

• But it's not very efficient in general (due to the constant rectification and re-inversion)

• And even less efficient in dealing with surges

• So it needs to be majorly oversized

• And thus costs a few thousand bucks.

• Ferroresonant models are more efficient

• But are freaking huge and also expensive.

So even if my apartment manager wants to charge me $500 for what ought to be a $200 job to run a new circuit, I'd have to move 5-10 times for a UPS to pay for itself, not including any electricity costs from the inefficiency, plus noise and heat...

I guess I'll talk to the apartment manager.

--
Jay Levitt                |
Boston, MA                | My character doesn\'t like it when they
Faster: jay at jay dot fm | cry or shout or hit.
http://www.jay.fm         | - Kristoffer

all that will do is correct for what the printer will see, you'll still see your lights dipping. What you need to do is measure the voltage at the panel before and after the printer is working. If you don't see any significant drop then you'll do fine with a new line to your printer how ever, if a drop is at the panel, you have bigger issues i'm sure your landlord will not want to deal with.

--
"I\'d rather have a bottle in front of me than a frontal lobotomy"
http://webpages.charter.net/jamie_5

--
How about if you made a mains "Y" adapter which necked down to a
single receptacle and you plugged the male ends into different
circuits on the same phase?

--
JF

Yeah, that (a ferroresonant transformer) might work, but then I'd have to rent another apartment to keep the transformer in :)

Interesting idea. Actually, in a previous apartment, I solved the problem by just running a 20A extension cord for the printer, plugged into the kitchen outlets. It may well have caused a voltage drop in the kitchen, but when I'm printing, I'm not in the kitchen. Therefore, I didn't care.

If the landlord doesn't want to run a new circuit - and now that I look at it, the printer's about 6 feet from the panel, with only a closet between them, so that's good - I could look into drawing power from a different circuit, either with said Y adapter or said "shift the problem" approach.

--
Jay Levitt                |
Boston, MA                | My character doesn\'t like it when they
Faster: jay at jay dot fm | cry or shout or hit.
http://www.jay.fm         | - Kristoffer

You're right, of course. I did that at my last apartment (built in 2000, so decent-but-not-totally-modern code) and it wasn't affecting other circuits, so I'm taking a wild guess that it won't here (2002 code) either. In fact, last time, my "office" bedroom had outlets on two different circuits, so I was able to arrange things so the printer was on one side and all the lights plugged into the other. Not so lucky here.

Unfortunately, my Fluke and my scope are buried in storage, my cheapo Home Depot meter can't show brief AC drops, and it'd be cheaper to add the circuit than to buy a proper meter to test it. But I should try the extension-cord test to see if the lights still dim when the printer's on a different circuit. Thanks for the idea.

--
Jay Levitt                |
Boston, MA                | My character doesn\'t like it when they
Faster: jay at jay dot fm | cry or shout or hit.
http://www.jay.fm         | - Kristoffer

"John Fields"

** Extremely dangerous suggestion !!

When one plug is removed from the outlet - it has LIVE pins.

Makes the parallel connected AC wall warts idea look tame.

....... Phil

"Jay Levitt"

** The light output from a CFL ( compact fluoro) is FAR less affected by a drop in the AC supply voltage than incandescent or normal fluoros - so try using them.

Your other idea is plain nuts.

....... Phil

cut There is a very nice industrial unit,(timer+relay), which is intended to limit inrush current, it puts a small resistor (1-10ohms)in series with the mains power when you switch on, and shorting that resistor out after about 100 milliseconds. That gives the powersupply a chance to slowly load the supply caps,(those are often the main reason for the powerdip), and avoid blowing fast fuses. It solved my problem, when about 6 computers were connected to a 500 Watts 12Vdc--->230Vac convertor. Even though total load was only about 250Watts, the convertor signaled overload. Adding the timer/relay solved all problems.