Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Currrent source 0 - 10 Amps

Status
Not open for further replies.

dwverzwy

New Member
I'm trying to build a current source for testing solid state overcurrent relays. At my disposal I have 0-120 variable AC and/or DC. 480 constant ac and 120 constant AC.

I'm pulling my hair out researching ways to do this. At my old job, technicians would build these as test carts with some types of transformers, but I've yet to find anything for this except a typical voltage variac.

I'm trying to get at least 0-10 amps adjustable.

Any ideas?
 
Use a fixed load? A variable voltage through a fixed load will give you adjustable current. Maybe they were using a Variance with a big load dump resistor?
 
Thanks for the reply. The test cart I'm working with does have a load bank...

so I=v/r a LOW resistance will give me the current I need.

so. 120 ac input and 10 ohms is 12 amps. I wonder if the resistor and fuses will hand this load...

If I remember correctly, the current sources that we used b/f were independent of voltage. In other words it was strictly currrent with very little potential and was quite safe even at high current.
 
If you want a custom load dump just use a bunch of 100 watt light bulbs in parallel.
Every 100 watt lightbulb you put in parallel will draw a little less than 1 amp. 12 amps at 120V AC is 1440 watts of power, I doubt the resistors can handle it.
 
Ok but I'm looking to minimize the watts/power. I want a strictly current output, if possible with minimal voltage.

It needs to be effective but somewhat presentable because it's for the test lab at work.
 
Depending on what the actual voltage drop across the device in test is you may only need to add a step down transformer to give you the low voltage and isolation at the same time.
If its just a current sensitive device like a circuit breaker or current relay you may only need a couple of volts at the output at most.
 
find a step down transformer with a low voltage higher amperage rating.

Being its going to be powering a near dead short get the lowest output voltage you can find.
I have seen ones that are rated at 1.6 volts 20 amps before so I know they are out there.
Any standard low voltage secondary transformer from a electronics supply catalog would be all you will need just make sure its secondary amperage rating is high enough.
Check out the ones used for powering the heater filiments on vacumm tubes. Some vacumm tubes only need a few volts but a fair amount of amps to heat them up.

If you only need 10 amps a standard 120VA transformer out of a 6/12 volt 10 amp battery charger would work well. just wire it up for the six volt input on the primary and then use one half of the seconday windings or seperate them from being a center tapped series connected 12 volt and put them in parallel so you have a 6 volt 20 amp output.

Then just run the whole thing off of the Variac to regulate the output current.
The variac will alow you to bring the voltage up from zero and all you will need to do is watch your amps to see what is going across your device under test.
Its the amp load that smokes transformers. Just make sure what you have is rated for more than what you are testing and you should never have any real problems.
 
Last edited:
It is exactly that, a current relay...

Any advice on how to get the high current at low voltage?

5V Regulated Power Supplies that put out 30A to 50A are cheap, **broken link removed**.

I recently posted a current regulator.

If you start with a 5V supply and use a big N FET as the regulator element, at 10A into a short, 5V is dropped across the FET, meaning that it will be dissipating 50W of power. If your heatsink is big enough, that is quite doable.

Write back if you need further help...
 
If you want a custom load dump just use a bunch of 100 watt light bulbs in parallel.
Every 100 watt lightbulb you put in parallel will draw a little less than 1 amp. 12 amps at 120V AC is 1440 watts of power, I doubt the resistors can handle it.
The trouble with that idea is that bulbs have a very high positive resistance coefficient.

Another option would be using 22µF capacitors or 330mH inductors.

You haven't said what the load is, this is assuming it's purely resistive, if it's reactive then it might interfere with the ballast reactor.
 
what you are trying to do will give misleading results since the majority of the wear of the contacts is due to breaking the high currents at high voltages causing contact arcing. This fact is reflected in relay ratings: most will say the contacts are rated for say 220VAC but only 30VDC.

If you want a real low power test and you are running DC through the relay you want to set up a flyback through the relay back to your DC rail allowing you to break the rated current at the application voltage while only dissipating the circuit losses.

Dan
 
Are you using the current sensing relays in a DC or an AC aplication?

It will help clarify what you are working with so we have a better idea of what you need.
 
Thanks for the help, a little more info

I am trying to build a current source for testing Overcurrent relays.
These relays have an input for three phase current. In the field, these inputs are taken from current transformers that step down current from the main bus to anywhere from 0 - 7+ amps. So the relay itself actually sees from 0-7+ amps. It is set to trip at some point in that range.

These inputs are non resistive, and only slightly inductive. There are internal coils inside the relay. I'm fairly certain their impedance is negligible. This is an AC application

So what I am looking for is a current source only. No voltage, no power. Strictly current. What I am trying to build is similar to a Doble or any other brand of relay test set without all the extra features.
 
If there's no voltage there's no current. You're trying to test a safety device in a situation which can't occur in the real world? What's the point of this test if the conditions under which it occurs can't be replicated in the environment it's going to be used in? Yeah it's a bit of a pain to set up but use a voltage source with a load dump to test it, or who knows what you'll be missing.

It's kind of like the listed MPG car makers post on the sticker. Doesn't mean much cause you can be certain of one thing, you'll never actually get that running the car because they're done without factoring in wind resistance and right from the output shaft of the motor without the transmission or drive train attached.
 
All theory aside I know exactly what your working with now! The voltage drops across what you are working with are usualy very small. A few tenths of a volt in some cases, And only a volt or two in others. I would still recomend the systen that I mentioned earlier. At the most you may need to add a 1 ohm resistor in series with the relay in test.
All you will need to do is watch your amps going through and do what adjustments are needed.

I cant think of any simpler or more basic way than that.

If you all ready have a variac all you will need is that low voltage transformer, and one of them can be found for less than $30
 
Last edited:
These relays have an input for three phase current. In the field, these inputs are taken from current transformers that step down current from the main bus to anywhere from 0 - 7+ amps. So the relay itself actually sees from 0-7+ amps. It is set to trip at some point in that range.

These inputs are non resistive, and only slightly inductive. There are internal coils inside the relay. I'm fairly certain their impedance is negligible. This is an AC application

So what I am looking for is a current source only. No voltage, no power. Strictly current. What I am trying to build is similar to a Doble or any other brand of relay test set without all the extra features.

Ok, then take a **broken link removed** or **broken link removed**, and then hook it up to **broken link removed** that has a 3 to 6V secondary rated at 10A. Hams call this a "filament" transformer, and they are readly available new or surplus. Connect a 0.5Ω 50W wire-wound power resistor in series with your Device Under Test (DUT) and a high-quality AC ammeter.

The variac will let you slowly ramp up the current through your DUT. The AC ammeter will show the current.
 

Attachments

  • DUT.jpg
    DUT.jpg
    86 KB · Views: 149
Thanks for everyone's help. I've got a day in the office tomorrow where I'll experiment with some of your ideas. Specifically, I'm going to run the 0-120vac variac through the power resistor bank that already exists on the test bench. I think that one of the resistors is 20 ohm so that would give 3 amps at 60 V. I just worry cause that's 180 watts, considerably dangerous to be playing with. After verifying that this works, I'm going to experiment with some of tcmtech and MikeMI's ideas.
 
I'd be more worried about the voltage, 60VAC will bite, especially from a variac as it isn't isolated from the mains.
 
I'd be more worried about the voltage, 60VAC will bite, especially from a variac as it isn't isolated from the mains.

That is why I suggested using the filament transformer. It isolates the DUT from the AC line. By stepping down the voltage, he can also utilize most of the range of the Variac, rather than just using the first few turns.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top