multimeter

[michaelahess]

DMM's may be more accurate but I like being able to see the "swing" in readings, escpecially if it's a true sine I'm looking at. I do a lot of car audio and DMM's aren't quick enough at displaying the constantly changing values in an analog amplifier stream.

 

[pengwin]

the main difference between the expensie ones and the cheapos are the that more expensive ones will measure true rms voltage values. If you are looking for accuracy make sure that you get one that reads true rms. You can google rms and find what i am talking about and what it means. I only have a few mins on here..... Fyi I am an Electrical engineer so I hope i know a bit of what i am talking about.

yeh, my older bro is an EE too (UIUC).

The only question i have is will this one.
http://www.radioshack.com/product/index.jsp?productId=2103176&cp=2032058.2032235.2032305&parentPage=family

work for testing PSU's. IE i wanna see if my Antec Pulls the advertised 17 amps.

 

[pcsteve]

Unfortunately (or not having seen a couple of the reviews as well lol), you will only be able to measure the dc voltage on this one as the dc current range maxes out at 10 Amps like we were talking about before. So being a digital meter, if the current flowing is too high, you'll either fry the guts, blow a fuse or the display will show an error of some sort. At least with an analogue meter you have chance to disconnect it before damaging it, as you have an instant indication that it's going full scale yeah?

 

[pengwin]

what about this one?

http://cableorganizer.com/triplett/digital-multimeter/digital-multimeter-dt9205a.htm

 

[Hicks121]

None of these meters will read AMPS!! You need a clamp on meter. & you are talking around $150 for a cheap one. I have a Fluke clamp-on meter & have checked my draw on my psu before, but it is tough being in cramped spaces in the case.

 

[pcsteve]

well, thats better, a 20Amp range for dc. This will be fine if you know the current draw is gonna be less than 20Amps (i.e. current rating of the 12v rail you wish to measure). Now and again, you'll come across a PSU that'll handle like 24 or 26 Amps or even more on the 12v rail (especialy on SLI compatible PSUs).... that's not to mention 30/40 Amps on the 5v rail. Multi 12v rail PSUs should in theory be ok because they don't just supply say 30Amps on one 12v rail. A dual rail PSU will prob. kick out something like 18Amps on 12v1 and say 14Amps on 12v2. So it all depends on a lot of variables.

 

[pcsteve]

Pengwin dude, you do realise that to measure current, you will have to break the circuit and insert the meter? As opposed to voltage measurements which are taken from a reference point i.e. ground or 0v.

 

[pcsteve]

None of these meters will read AMPS!! You need a clamp on meter. & you are talking around $150 for a cheap one. I have a Fluke clamp-on meter & have checked my draw on my psu before, but it is tough being in cramped spaces in the case.

Why will he need a clamp on meter 8O I think, from what he said earlier, he doesn't want to pay an arm and a leg for one, and these meters will measure current. :roll:

 

[pengwin]

ok, so if i wanna measure voltage. then any cheapo one will do?

please bear with me, i have never seen or used a multilmeter in my whole life.

thanks

 

[Hicks121]

Any cheap one will do then, but remember, you do get what you pay for.

 

[Whizzard9992]

Check e-bay.

FLUKE 73III Digital MultiMeter 73 III Multi Meter

FLUKE 73III Digital MultiMeter 73 III Multi Meter NEW

Fluke Macto Tools MD73AS3 Handheld Multimeter No Res

Get a used one for what you're doing. Flukes last a lifetime.

ALSO: Before you get a radio-shack multi-meter, CHECK THE RATINGS. Some multi-meters are great for testing 5v and 12v circuits for HOBBY EQUIPMENT at 500ma. When you're talking 500W, though, you better have a meter that can handle it or you're going to melt the thing.

I have a $10 radio-shack mm at home, cuz I work with all low-voltage CMOS. I have a fluke at work because sometimes I need a little more oomph (testing [home-grown] power supplies and such).

 

[sirheck]

hey penguin any meter will do fine except for the amps
i have a snap on <fluke> it was 600 dollars
and i bought a clamp on amp probe for 200
also a snap on but it was 175 dollars
and it goes to 10 amps max dc only
snap on does make one that goes to
1000 amps ac or dc but it is around 250 dollars
i dont think you want to spend that much?
i cant think of any good and cheap meters
with an amp rating of over 10 amps.

 

[techgeek]

I am an Electronics Engineering Technologist, at work I use Fluke. I have a Fluke 87 (excellent multimeter and measures capacitance), it is fused for current at 470mA and 10A. That said I believe it would be overkill for you to get one, also probably out of your budget.

If you want to measure a PSU's current output on a given rail, jump start it (outside of a system) and place a power resistor on that rail to ground (of course you'll have to use Ohms Law to find the value of resistor to get 17A from your 12V rail which by the way is .7 ohms) then measure the voltage across the resistor. If it maintains 12V within the tolerances laid out by the ATX specifications and can sustain it over a given amount of time you will know that it is indeed putting out 17A. Unfortunately the resistor will have to have a power rating in excess of 200W, good luck finding one of those. The other option would be to use several resistors in parallel to simulate the same load resistance, this way you reduce the needed power rating for your test resistors. So with the same example you could use 3 X 2.1 ohm resistors thereby reducing the needed power rating to 66W. This is getting closer to realistic power ratings, you may have to increase the number of resistors in parallel to get a lower power rating. There is a formula for calculating resistor in parrallel.

Not only will this tell you whether the supply can provide the rated current but it will tell you whether or not it can maintain it. If you wanted to go one step further you could use an oscilloscope to measure ripple @ full load to see if it stays withing the ATX specs.

You can see why profession PSU testing equipment is so expensive (also more sophisticated than this as they use active components to simulate load changes etc).

EDIT: I just read that you are testing to see if your PSU will put out the published current. My procedure is the only way you can do it, you can't do it within you system. This is because there is no way to put your PSU under a consistant load to get the maximu current specified my Antec. The only way to do it is to do it in a controlled environment like I suggested. Or you could see if anyone has done this with your supply (read Tomshardware PSU Stress Tests), unless you want to know if your specific supply meets the specified ratings.

 

[pengwin]

good ness.


ok im gonna get the 20 dollar one at cable organizer.com

 

[pcsteve]

Good shout mate, especially if it's to monitor your 12v 17A, cos it covers that anyway. Good luck man

 

[pengwin]

Good shout mate, especially if it's to monitor your 12v 17A, cos it covers that anyway. Good luck man

nha im prolly gonna get bigger PSU's but i'll worry about it then.

 

[Whizzard9992]

I am an Electronics Engineering Technologist, at work I use Fluke. I have a Fluke 87 (excellent multimeter and measures capacitance), it is fused for current at 470mA and 10A. That said I believe it would be overkill for you to get one, also probably out of your budget.

If you want to measure a PSU's current output on a given rail, jump start it (outside of a system) and place a power resistor on that rail to ground (of course you'll have to use Ohms Law to find the value of resistor to get 17A from your 12V rail which by the way is .7 ohms) then measure the voltage across the resistor. If it maintains 12V within the tolerances laid out by the ATX specifications and can sustain it over a given amount of time you will know that it is indeed putting out 17A. Unfortunately the resistor will have to have a power rating in excess of 200W, good luck finding one of those. The other option would be to use several resistors in parallel to simulate the same load resistance, this way you reduce the needed power rating for your test resistors. So with the same example you could use 3 X 2.1 ohm resistors thereby reducing the needed power rating to 66W. This is getting closer to realistic power ratings, you may have to increase the number of resistors in parallel to get a lower power rating. There is a formula for calculating resistor in parrallel.

Not only will this tell you whether the supply can provide the rated current but it will tell you whether or not it can maintain it. If you wanted to go one step further you could use an oscilloscope to measure ripple @ full load to see if it stays withing the ATX specs.

You can see why profession PSU testing equipment is so expensive (also more sophisticated than this as they use active components to simulate load changes etc).

EDIT: I just read that you are testing to see if your PSU will put out the published current. My procedure is the only way you can do it, you can't do it within you system. This is because there is no way to put your PSU under a consistant load to get the maximu current specified my Antec. The only way to do it is to do it in a controlled environment like I suggested. Or you could see if anyone has done this with your supply (read Tomshardware PSU Stress Tests), unless you want to know if your specific supply meets the specified ratings.

lol. I'm calling shananigans (sp?) on this one.

17A @ 12v won't evaporate, and there's not a resistor/resistor array on the planet that will make electricity vanish. The only thing you can do is convert it, and what does electricty like to be converted to? Heat. So unless you want to jimmy-rig a 500W heater, I suggest you ignore the above comment.

If you push your PSU beyond its limitations, stated or not, one of 3 things will happen:

1) Your rails will begin to droop
2) You'll blow a fuse
3) Your PSU will catch on fire

So if you want to see if your PSU can handle it, the easiest way is just to check the voltage at full load. If your PSU isn't getting hot and your voltages are reading A-OK, then be happy

In short, The Radio-Shack DMM should do fine to check your rails. Just run 3dmark or something while you take samples.

 

[Whizzard9992]

The more I read your post, the more I laugh. Thanks!

I'm an Electronics Engineering Technologist

I work with a LOT of EE's, and not a single one refers to themselves as a "Electronics Engineering Technologist". It sounds pretty gay to me, I'm not gonna lie.

And one more thing, lol. Resistors don't "load" a circuit, they merely constrict it. You still need something drawing power. What you're talking about is creating a short-circuit, which is going to catch fire unless you deal with a LOT more than 3 x 2.1 ohm (ROFL) resistors.

A real EE would know that.

 

[Batspoon]

lol. I'm calling shananigans (sp?) on this one.

17A @ 12v won't evaporate, and there's not a resistor/resistor array on the planet that will make electricity vanish. The only thing you can do is convert it, and what does electricty like to be converted to? Heat. So unless you want to jimmy-rig a 500W heater, I suggest you ignore the above comment.

If you push your PSU beyond its limitations, stated or not, one of 3 things will happen:

1) Your rails will begin to droop
2) You'll blow a fuse
3) Your PSU will catch on fire

So if you want to see if your PSU can handle it, the easiest way is just to check the voltage at full load. If your PSU isn't getting hot and your voltages are reading A-OK, then be happy

In short, The Radio-Shack DMM should do fine to check your rails. Just run 3dmark or something while you take samples.

The more I read your post, the more I laugh. Thanks!


I'm an Electronics Engineering Technologist

I work with a LOT of EE's, and not a single one refers to themselves as a "Electronics Engineering Technologist". It sounds pretty gay to me, I'm not gonna lie.

And one more thing, lol. Resistors don't "load" a circuit, they merely constrict it. You still need something drawing power. What you're talking about is creating a short-circuit, which is going to catch fire unless you deal with a LOT more than 3 x 2.1 ohm (ROFL) resistors.

A real EE would know that.

You're dead wrong...


First, 17 Amperes on 12 Volts DC draw about 204 Watts. Not 500...

Second, you can find 200 W single resistors and more Aluminum Wirewound resistors for less than 30 $, but if you're really on a low budget... the resistor network (8 to 10 res in that case) may get more expensive than your digital multimeter...

With a 0R77 (0.77 ohms) load on 12v DC you will get 15.58 Amps.

You can get 0R47 and 0R1 200W resistors easily, and doubling the "network" in parallel you can get a 400W psu load simulator.

0.47 + 0.1 + 0.1 + 0.1 (200 Watts resistors)
0.47 + 0.1 + 0.1 + 0.1
___________________
0.77 Ohms 400W


0R47 (0.47 Ohms) 200 Watts Wirewound Aluminum Clad Resistor

0R1 (0.1 Ohms) 200 Watts Wirewound Aluminum Clad Resistor

It's better to attach them on a big aluminum surface or on a efficient heatsink as they can go to 250ºC max @ full load, in that case 400W (with 200 W resistors in parallel), obviously impossible to obtain with 0.77 Ohms on a 12 Volts rail. Remember to double the network in parallel !!!



Disclaimer, Pengwin if you aren't sure or comfortable with don't play with that kind of things.

If the links don't get you on the product page, enter the United Estates Export (at bottom) in the Farnell's site and use the product code 272670 for the 0R47 resistor and the 272668 for the 0R1 one. Or just PM me.

 

[Batspoon]

And to be sure, I would go with 0R87 or 0.87 Ohms at first (you need to be aware of the 5% tolerance for this particular resistors), and them take off 0.1 Ohm from each serial resistor line, measure the ressistance with a dmm and if it don't read less than 0.77, go ahead and stress your psu.

I did made a simple diagram:

Of course you can try to find smaller resistors like 0.05 Ohms to adjust the load more. Also it's a good think to do to search for 50W resistors (cheaper than 200W, obviously) and make a 7 serial lines in parallel network for a total power draw of 350 W. Maybe this way you can fit your budget.

 

[techgeek]

The more I read your post, the more I laugh. Thanks!


I'm an Electronics Engineering Technologist

I work with a LOT of EE's, and not a single one refers to themselves as a "Electronics Engineering Technologist". It sounds pretty gay to me, I'm not gonna lie.

And one more thing, lol. Resistors don't "load" a circuit, they merely constrict it. You still need something drawing power. What you're talking about is creating a short-circuit, which is going to catch fire unless you deal with a LOT more than 3 x 2.1 ohm (ROFL) resistors.

A real EE would know that.

Yes a resistor from rail to ground will provide a load. Not only a load but a constant load. Three 2.1 ohm resistors in parallel make a .7 ohm load. "Constrict"? do you mean resist, thus the name "resistor". Apparently I'm dealing with someone who has no idea about voltage current, resistance or power. 12V across a .7 ohm load makes a current draw of ~17A. This is fundamentaly Ohms Law. I can tell you looking in my ElectroSonic, Newark and Allied Electronics catalogs, there aren't too many 200W power resistors, and they are expensive. Using a many resistor network in parallel is more practical. As for heat dissipation, the power resistors that I prefer are aluminum chassis mounted. These can be mounted to a metal plate to dissipate heat. I have a resistor network I built as a load simulator in my lab with me as I type this. It uses 50W power resistors, but since it's not meant to check PSU (industrial purposes not commercial), they are a much higher resistance than would be needed to conduct the experiment Pengwin wishes to conduct.

As for being a EET, that's what I am. Graduated 94 from DeVry. Working in the field since graduation with the same company. Competent enough to be working abroad managing my own lab. Thank you very much.

 

[techgeek]

What you're talking about is creating a short-circuit

You figure. How else do you propose to get 17A from a 12V rail. Simple math says if the current require is more than the voltage, obviously the resistance is going to be less than 1 ohm. I can assume you can manipulate an equation such as this:

V=I x R where V=voltage I=current and R=resistance.

I can't make it any plainer than that for you. Notice I used capital letters in the formula, that's because we are talking in terms of DC (Direct Current) had we been talking about AC, I would have use lower case. Be glad this is DC and not AC, or we would have to be talking about reactance whereby you can have a leading or lagging load (capacitive or inductive)

So in essense you could say we are creating a short, if your definition of a short is less than an ohm. Many loads (including this example) come very close to being a short, but the power supply is designed to supply the current required by that load. If that blows your mind, use the equation to figure out what sort of load you need to get a 30A draw from a 5V rail (which is a typical current supplied by most ATX supplies). Just in case you can't manage a calculator it's .1666 ohms. $hit don't do that is will blow up your supply :roll: . If your PSU is rated for it, it should be able to do it.

I won't lie to you testing PSU is a complicated business, and if you've read Tom's PSU Stress Test, you will see that there are no real standards for deriving maximum current output. You will also discover how many supplies that people believe are good, don't pass.

Sorry if I got a little hot under the collar, but when someone tells you that you don't know what you're talking about when you obviously know that you do, it tends to make you a little upset.

 

[Whizzard9992]

As for being a EET, that's what I am. Graduated 94 from DeVry. Working in the field since graduation with the same company. Competent enough to be working abroad managing my own lab. Thank you very much.

Yeah I'm not buying that.

Spouting off simple equations doesn't prove your point, it doesn't discredit me, and most importantly it doesn't at all address the issue at hand.

All the guy wants to do is test his PSU. He's talking about a $20 device. You're talking about building a resistor array that's going to heat to 500 degrees. Yeah ok. He doesn't know what a multi-meter is, and you want him to short-circuit his power-supply. Good idea :roll:

Obviously education doesn't imply competence.

All he has to do is load the PSU with what he has already, and test the rails. The Amperage rating is based on the quality of the components and the type of circuit. Anything above the rating should blow the fuse, and below that it should operate normally.

Oh and by the way, shorts cause heat across the entire circuit, so chances are he's going to blow a fuse in his PS or catch his PSU on fire. Good call Mr. Electronics Engineering Technologist.


Pengwin, just get a DMM. Any $20 one should be fine. There are a lot of tutorials online. If you pushed the PSU to the max, you'd prolly just blow a fuse. You don't want to do that. Just test the rails with a DMM while running 3dmark. There's no better way to do it without some expensive equipment or some (real) EE knowledge.

 

[techgeek]

I didn't say I think he could do it, I was trying to illustrate to him that finding out if his supply is able to put out the rated max would take more expertise than he has. Testing PSU's is a very precise and complex undertaking.

He said he wanted to test if it could put out the maximum, there is no way to do that in his system, you just can't put a constant load on the PSU to run it at it's maximum. Even if his system could, that would mean that his PSU was severely underpowered for his system, you have to have some overhead because you never want you PSU operating at it's maximum.

You can believe what you like about me, far be it from me to convince a forum troll that I am what I say I am. You obviously are not someone who has any idea about electronics. My credo: Never argue with an idiot, people watching may not know the difference.

As for Pengwin, I honestly think the experiment that you've set for yourself is neither worth the time, nor within your expertise. That's not an insult, just honesty. If you are only concerned with current draw in you system, forget about directly measuring current. Just measure you rails while under max load. It's just not feasible for you to try and bring your PSU's rail(s) to maximum current output just to see if it can do it. You didn't actually mention why it is that you want to do this. Is it because you think you PSU is underpowered for you system? Are you experiencing lock ups or re-boots?

 

[techgeek]

Thank you Batspoon for coming in and backing me up. Unfortunately we have not educated the young Wizard or whatever his name is. Nice touch with the schematic, what did you use? I have experience with Orcad (when I was still going to school), but I couldn't tell what you used from the schematic you provided.

Edit: I just went back over your posts Pengwin, and I may have misunderstood you, and you may have a misconception of how your PSU works. You said that you wanted to see if it was putting out the max rated 17A while in your system. The only way this would happen is if the load required it, it doesn't put the max rated current out all the time, only when you system needs it to. If in fact it ever does reach this limit, your PSU is definitely underpowered for your system.