How to prevent increasing current above a certain level?

[benofer90]

Any one knows a simple way to prevent an increasing current from going above a certain value?
Current is increasing over time. Only when it reaches a certain value then I would like to stop it from going above and keep it at that Max value .

thank you.

 

[anorlunda]

I googled current limiting and found this

https://6002x.mitx.mit.edu/wiki/view/Current-andVoltage-limitingResistors

 

[axemaster]

If you want a useful reply you'll need to tell us more about what you're doing. What are you building, how big of a current are we talking about? What will this be used for?

 

[berkeman]

Any one knows a simple way to prevent an increasing current from going above a certain value?
Current is increasing over time. Only when it reaches a certain value then I would like to stop it from going above and keep it at that Max value .

thank you.

If you want a useful reply you'll need to tell us more about what you're doing. What are you building, how big of a current are we talking about? What will this be used for?

Exactly. Can you please post a schematic for what you have already? There are lots of different current-limiting techniques that apply in different situations.

 

[meBigGuy]

How rapidly does the current rise?
Can you tolerate any overshoot?
How much current are you talking about?
How accurate must the limit be?
What voltages are involved?
What controls do you have over the source of the current?
plus a half dozen other questions I don't know to ask yet.

The quality of the answers on this forum are highly correlated with the quality of the questions! <-- maybe that's my new sig

 

[benofer90]

Hi meBigGuy, berkeman, axemaster, anorlunda

Answering some of your valid questions below:

QUESTION : If you want a useful reply you'll need to tell us more about what you're doing. What are you building, how big of a current are we talking about? What will this be used for?
ANSWER: i am building a DC generator, thus the current increase with relation to the RPM

QUESTION : How rapidly does the current rise?
ANSWER: it can rise fast if i go from 10 RPM to 10,000 RPM in short period of time

QUESTION : Can you tolerate any overshoot?
ANSWER: i believe the answer is yes , I just don't want to fuse the cooper wire. I am using AWG #5 which can handle a load of about : 60/75/90 °C => 55 / 65 / 75 (A)

QUESTION : How much current are you talking about?
ANSWER: I want to have the option to go as high as 40 but aiming for 5-20. it will be a set number once i decide . meaning it won't be a range unless a range is simple to implement .

QUESTION : How accurate must the limit be?
ANSWER: it doesn't need to be accurate. if my max is 10 Amps then it can still be 8 or 12

QUESTION : What voltages are involved?
ANSWER: it is really function of RPM at 10,000 RPM voltage can be 70

QUESTION : What controls do you have over the source of the current?
ANSWER: full control as its in development , but still dealing with magnets which are constant .

QUESTION : plus a half dozen other questions I don't know to ask yet.
ANSWER: Please ask if needed

QUESTION : Exactly. Can you please post a schematic for what you have already? There are lots of different current-limiting techniques that apply in different situations.
ANSWER: its a straight forward generator . magnets, coils and copper wire .

thank you all again

 

[berkeman]

Thanks for the extra details -- those help a lot.

I would just use a commercial off-the-shelf circuit breaker. If there is an overcurrent, the breaker trips and needs to be reset manually after the too-heavy load is removed.

 

[benofer90]

Thanks for the extra details -- those help a lot.

I would just use a commercial off-the-shelf circuit breaker. If there is an overcurrent, the breaker trips and needs to be reset manually after the too-heavy load is removed.

Negative , it will turn off the entire thing . no can do .

 

[berkeman]

Negative , it will turn off the entire thing . no can do .

Roger that. So you want a current limit that will drop the output voltage when the load gets too big. That's fairly easy using a couple transistors and a resistor. Your currents and voltages are pretty large, but still do-able. This generator is DC or AC?

 

[berkeman]

If it's DC, you can use a circuit like this. You size the sensing resistor to drop around 0.6V at maximum current (there is a temperature coefficient that comes into play...):

http://forums.parallax.com/uploads/attachments/58478/86715.gif

 

[benofer90]

Roger that. So you want a current limit that will drop the output voltage when the load gets too big. That's fairly easy using a couple transistors and a resistor. Your currents and voltages are pretty large, but still do-able. This generator is DC or AC?

yea thanks. that sounds like a plan. DC generator. remember i only want to limit it once it reaches the require amp. let say 10 amps . but as the RPM goes to 10,000 the amount of AMPS and VOLTAGE will try go higher even though we are limiting it. so the system will get hotter i assume . i just don't want to fuse or melt the wires and their insulation .

so where do i start ?

 

[axemaster]

There are a few general ways to go about this...

1. Control the speed of the generator. This is the best method, as it doesn't require dissipating huge amounts of heat.

2. Use opamp controlled transistors to control the current. The transistors will have to operate in their linear region (i.e. partially on), meaning they will be highly resistive and will generate a LOT of heat. Based on your description, the heat will be hundreds of watts, so thermal design will be very important and the transistors will have to be balanced. This is probably way beyond what you'll be able to do with your current knowledge.

3. Use a current controlling buck converter. This has the advantage of operating MOSFETs in their full on/off states, thus keeping heat to a minimum, and is highly efficient. I imagine 96% or higher efficiency would be easy to achieve in your system. But again, this will require knowledge you don't have yet.

In general I would be very cautious in experimenting with such a high power system. 70V is on the edge of what I'd be comfortable with, and the fact that there are 100's of watts waiting to be unleashed means that it would be easy to melt metal or cause something to explode with great violence. Additionally, you might have inductive voltage spikes coming out of the generator that will wreck your components.

No offense, but if you have to ask this question, you shouldn't be trying your hand on such a high power system. If you're serious about doing this, you should get an experienced electronics engineer to do it for you.

 

[benofer90]

There are a few general ways to go about this...

1. Control the speed of the generator. This is the best method, as it doesn't require dissipating huge amounts of heat.

2. Use opamp controlled transistors to control the current. The transistors will have to operate in their linear region (i.e. partially on), meaning they will be highly resistive and will generate a LOT of heat. Based on your description, the heat will be hundreds of watts, so thermal design will be very important and the transistors will have to be balanced. This is probably way beyond what you'll be able to do with your current knowledge.

3. Use a current controlling buck converter. This has the advantage of operating MOSFETs in their full on/off states, thus keeping heat to a minimum, and is highly efficient. I imagine 96% or higher efficiency would be easy to achieve in your system. But again, this will require knowledge you don't have yet.

In general I would be very cautious in experimenting with such a high power system. 70V is on the edge of what I'd be comfortable with, and the fact that there are 100's of watts waiting to be unleashed means that it would be easy to melt metal or cause something to explode with great violence. Additionally, you might have inductive voltage spikes coming out of the generator that will wreck your components.

No offense, but if you have to ask this question, you shouldn't be trying your hand on such a high power system. If you're serious about doing this, you should get an experienced electronics engineer to do it for you.

thank you axemaster, yes the RPM limiting Idea is part of what I was thinking on doing . i think maybe better for me to limit the RPM and by that control the entire thing . but still i want to limit the current so i will use ur suggestion and "berkeman's"

I guess I need an RPM sensor any idea?
then i can just put a stop on the entire thing .
also I am not planing on going more then 500 RPM which will give me 7V

 

[berkeman]

3. Use a current controlling buck converter. This has the advantage of operating MOSFETs in their full on/off states, thus keeping heat to a minimum, and is highly efficient. I imagine 96% or higher efficiency would be easy to achieve in your system. But again, this will require knowledge you don't have yet.

Great idea!

 

[axemaster]

Obviously I don't know your system, but to measure RPMs you could use a few things:

Encoders - This attaches to the spinning shaft. It's basically a ring with holes in it, and a light that shines through the holes. The frequency of the light flashing tells you how fast the thing is going. You can do a similar thing with magnets and hall effect sensors, or with capacitive sensors, or so on. Basically anything that converts the rotational speed into a frequency that you can measure. Also, you might be able to listen to the generator with a mic and calculate the speed from that.

500 RPM which will give me 7V

This may be the case at steady state. But when the current is changing, there is the possibility of creating voltage spikes. This is because the current has "momentum". If the generator is supplying 10A and you unplug the load, the output voltage will momentarily spike as it tries to keep going.

If you use Berkeman's method, just be aware that you may have to dissipate very large amounts of heat. This is not trivial.

 

[berkeman]

If you use Berkeman's method, just be aware that you may have to dissipate very large amounts of heat. This is not trivial.

I want to change my answer!

 

[axemaster]

I want to change my answer!

The buck converter came to mind because recently I was designing one to handle several kW... with a $2 comparator and 2 medium size MOSFETs. At 99.8% efficiency. It didn't even need a heat sink. That was the moment when I decided I like switchers.

That said, switchers have the disadvantage of being slow to respond and not being "real" regulators. So your idea was perfectly valid.

 

[jim hardy]

Controlling an old fashioned generator ? Hooowww absolutely Delicious !

Does it have a field?
Great time to learn generator basics.

You can still buy electromechanical voltage regulators. They are really fun, you can push gently on the relays and watch the system work.
Makes the DC machine intuitive.

tutorial on how they work:
(^^^^^That picture comes from here-)
http://www.stinsonclub.org/PublicTech/YahooGroup/Delco-Remy 1R-116 (regulator).pdf

 

[Jeff Rosenbury]

There exist resistors which are very temperature dependent. They are used in ceramic heaters to limit the current (and the temperature). As the current rises it will heat the resistor, raising the resistance. It may be a simple solution?

 

[benofer90]

Obviously I don't know your system, but to measure RPMs you could use a few things:

Encoders - This attaches to the spinning shaft. It's basically a ring with holes in it, and a light that shines through the holes. The frequency of the light flashing tells you how fast the thing is going. You can do a similar thing with magnets and hall effect sensors, or with capacitive sensors, or so on. Basically anything that converts the rotational speed into a frequency that you can measure. Also, you might be able to listen to the generator with a mic and calculate the speed from that.This may be the case at steady state. But when the current is changing, there is the possibility of creating voltage spikes. This is because the current has "momentum". If the generator is supplying 10A and you unplug the load, the output voltage will momentarily spike as it tries to keep going.

If you use Berkeman's method, just be aware that you may have to dissipate very large amounts of heat. This is not trivial.

so i am finding out that RPM sensor is ferry easy to use , yet the limiting part is bit tricky . so i know the the RPM or i get a pulse that i can count but i need another component with a Max switch to do something(for example cut the current) at X RPM.

 

[benofer90]

Great idea!

Hi axemaster , berkeman
I think it might work but still not sure what will happen if the V is lower then the requirement and if the Amps exceeds the limit.

and does it have an On/Off switch i can use at X current . ?

For Example :
http://www.bambuzo.com/us/yeeco-dc-...gulator.html?gclid=CN2AkaXI2McCFdIXHwodHx4DoQ

http://www.gearbest.com/development-boards/pp_140282.html?currency=USD&gclid=CNmE6OjC2McCFUWRHwodwmoEkQ

 

[benofer90]

There exist resistors which are very temperature dependent. They are used in ceramic heaters to limit the current (and the temperature). As the current rises it will heat the resistor, raising the resistance. It may be a simple solution?

would you be able to direct me to a link or do you have the name of it ? and can i have an On/Off switch at X current ?

 

[benofer90]

Controlling an old fashioned generator ? Hooowww absolutely Delicious !

Does it have a field?
Great time to learn generator basics.

You can still buy electromechanical voltage regulators. They are really fun, you can push gently on the relays and watch the system work.
Makes the DC machine intuitive.View attachment 88118

tutorial on how they work:
(^^^^^That picture comes from here-)
http://www.stinsonclub.org/PublicTech/YahooGroup/Delco-Remy 1R-116 (regulator).pdf

Hi Jim, is there a modern version assembly of the same kind?
will it limit the current?
does it have an on/off switch at X current?

 

[jim hardy]

Hi Jim, is there a modern version assembly of the same kind?
Similar but not exact. Fords of early 70's are probably your best bet.
will it limit the current?That one actively limits armature current by modulating field current. Middle coil does that.
As automakers replaced generators with alternators, then got better with alternator design, it became unnecessary to limit armature current and that coil disappeared from regulators. Alternators are inherently self limiting because of something called "Armature Reaction:". That might be an approach for you to look into.
Alternators, having diodes, no longer needed that reverse current cutout relay either. My 1962 Chrysler had a one coil voltage regulator, as i said Ford regulators s were more complex. and i think limited current for a few years.
Point being - there is a window where you might find a more modern regulator made for alternators that'll still regulate current.
does it have an on/off switch at X current? No, it disallows more than X current by reducing field excitation . Only turn-off function is leftmost coil which disconnects in event current commences flowing into rather than out of the generator. An electromechanical diode, if you will.

old style 3 coil regulators are available brand new but might cost you $50.
I;ve never looked into modern solid state ones because they're not repairable, just replace. Other posters here have suggested approaches to make one, and their skills vastly exceed mine..
If you just switch off at X current, current immediately falls to zero - will you prevent an automatic "switch back on" ? Don't build yourself an oscillator.

 

[axemaster]

@benofer90

You might be able to use the buck converter you mentioned at:
http://www.gearbest.com/development-boards/pp_140282.html?currency=USD&gclid=CNmE6OjC2McCFUWRHwodwmoEkQ

It claims to be current limiting at 10A. Whether this is true or not... well, without a datasheet, who can say? Also, it says that the max input voltage is 36V, so if your generator exceeds this, expect trouble.

Looking at the bottom of the board, it appears to have 2x MOSFETs, which probably means it's high efficiency (synch rectification). The presence of a capacitor connected to one of the legs of the leftmost mosfet also suggests bootstrapping, which is good.

No, these devices don't seem to have on/off switches. I don't know what it will do if under voltage, that's something only the datasheet could tell. Presumably it turns off though.

In general I would beware of buying anything without good documentation. It's not that the product is crap - it's that you need to understand how it works in order to use it effectively.

 

[Hesch]

If the magnets in the motor are electromagnets, why not put the rotor current through some windings around these magnets. Thus an increasing rotor current will demagnetize the magnetic field ( if the rotor current is passing the right direction through ). It will be some kind of a traction motor:

No heat, no energy losses, fully automatic, maybe sufficient precision ?

 

[Jeff Rosenbury]

The buck converter came to mind because recently I was designing one to handle several kW... with a $2 comparator and 2 medium size MOSFETs. At 99.8% efficiency. It didn't even need a heat sink. That was the moment when I decided I like switchers.

That said, switchers have the disadvantage of being slow to respond and not being "real" regulators. So your idea was perfectly valid.

How do switchers work with electric machines? Is there a problem with inrush current?

 

[benofer90]

If the magnets in the motor are electromagnets, why not put the rotor current through some windings around these magnets. Thus an increasing rotor current will demagnetize the magnetic field ( if the rotor current is passing the right direction through ). It will be some kind of a traction motor:

No heat, no energy losses, fully automatic, maybe sufficient precision ?

maybe in other generators . can not use it in mine. but thanks for your idea.

 

[axemaster]

How do switchers work with electric machines? Is there a problem with inrush current?

That would depend on the machine, but generally speaking they have a more gentle startup than linear regulators. So it shouldn't be an issue.

 

[benofer90]

@benofer90

You might be able to use the buck converter you mentioned at:
http://www.gearbest.com/development-boards/pp_140282.html?currency=USD&gclid=CNmE6OjC2McCFUWRHwodwmoEkQ

It claims to be current limiting at 10A. Whether this is true or not... well, without a datasheet, who can say? Also, it says that the max input voltage is 36V, so if your generator exceeds this, expect trouble.

Looking at the bottom of the board, it appears to have 2x MOSFETs, which probably means it's high efficiency (synch rectification). The presence of a capacitor connected to one of the legs of the leftmost mosfet also suggests bootstrapping, which is good.

No, these devices don't seem to have on/off switches. I don't know what it will do if under voltage, that's something only the datasheet could tell. Presumably it turns off though.

In general I would beware of buying anything without good documentation. It's not that the product is crap - it's that you need to understand how it works in order to use it effectively.

Yes I'm going to ask for the data sheet before moving forward with anything .
i did see these online can any of them work ? do they need a permanent V to operate ?

http://www.ebay.com/itm/1-Channel-DC-5v-AC240V-DC30V-10A-Relay-Interface-Board-Normally-Open-Module-/111420751580?hash=item19f131bedc

http://www.ebay.com/itm/One-Channel-DC-12V-Coil-10A-250V-125VAC-30V-28VDC-Relay-Module-Board-/311296747183?hash=item487abb66af

http://www.ebay.com/itm/10-2N6111-30V-7A-PNP-Switching-Transistor-NEW-/400435422932?_trksid=p2141725.m3641.l6368

https://www.google.com/shopping/pro...d=0CJQCEPMCMARqFQoTCKyY6YrW2ccCFUxZPgodp-AFAA

i do need to have an automatic on/off switch of some kind at a specific current . if the current is under X then it should stay ON and if above then should be OFF

any idea?