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.

how to best limit mains current?

Status
Not open for further replies.

user1453

New Member
I have a load(inductive) that won't be able to limit it's own current.

I have attempted a current limiter using an lm317 but when I ajust the voltage over 100v the circuit stops working

what I want to end up with is a current limiting circuit that will work in the 0-300v range and allow regulatate the current to be less than a preset value. The range of this being 1-8A

there is a H-bridge around the inductor already, I just want to limit the current entering the h-bridge

Anyone hae any ideas?
 
Last edited:
300v at 8A is a lot of power, building a precise control for that would be difficult. Why does the input voltage have to cover such a huge range ? Zero included.. well good luck with that one :)

The topic mentions mains, but you talk of H bridges and 317's like it's DC. Is it AC or DC you need to limit ?
 
Last edited:
The H-bridge will be fed from rectified smothed mains 240V AC

the current needs to be limited in order to prevent the mosfets from being damaged

If I use a pwm with deat time to limit the current then freewheeling current will damage the fast recovery diodes in antiparalell with the mosfets

the voltage will eventually be rectified mains voltage 240V

however, initially I would like to test the circuit at a lower voltage probably 60v to prove the circuit
 
well simple solution then

1) rectify the AC (to ~350Vdc)
2) have a nice biig DC-link capacitor (if really big check for in-rush current)
3) use IGBT's (FET's no good over 100Vdc)
4) use IGBT's in H-bridge configuration (NOTE IGBT's do not have intrinsic diodes so either get ones that have anit-parallel diode or add diodes)
5) have the inductive load at the H-bridge terminals.
6) since current ~8A look into hall effect sensor and put that in series with load
7) make a simple current controller with a demand that is variable for yr 1-8A (and feedback from hall-effect)
8) a P-controller should suffice since DC/low-bandwidth
9) this feeds PWM generation stage (since 8A can afford to get to switching freq ~50kHz)
10) since decent DC-link potential yr IGBT gate-drive will need to be isolated/floating
 
Last edited:
the H-bridge frequency must be independant of the current
the circuit will be used to investigate reesonance in transformers

My problem is I haven't found a circuit that will work at up to 350V
the current limiting is to limit the total power into the system and hoperfully reduce heating, before the circuit is run it will be set at a value between 1-8A

the control circuitry for the pwm signal and gate drive is already designed,

all I need to do now is find a way to limit the current into the H-bridge
 
user1453 said:
the H-bridge frequency must be independant of the current
the circuit will be used to investigate reesonance in transformers

My problem is I haven't found a circuit that will work at up to 350V
the current limiting is to limit the total power into the system and hoperfully reduce heating, before the circuit is run it will be set at a value between 1-8A

the control circuitry for the pwm signal and gate drive is already designed,

all I need to do now is find a way to limit the current into the H-bridge

And I told you how
*Switching freq is independant of the current, it is derived from a fix-freq triangle-wave at the PWM-stage

*The only difference between a H-bridge that works at 12V and a H-bridge that works at 350V is the silicon (and I told you to use IGBT's that have blocking potential of 600V,1200V,1700V dependant on what chosen) and also the gate-drive (which I siad need to be isolated & floating, ie use opto's and isolated DC:DC converters)

*since the load is inductive the only power will be in the losses, there is a high chance that your PWM-duty will be around 5~10%

* since you say you have the gate-drive (they are isolated?) and the PWM stage, all you need is a current-loop to drive the PWM-stage

1) a differential-amplifier that has as its inputs
a) current demand
b) current feedback (which comes from a hall-effect in hte current-path like I said)

with the current-error derived (feeding a P or PI gain-block), this then feeds to yr PWM stage that should be a comparator
 
user1453 said:
I have a load(inductive) that won't be able to limit it's own current.
What sort of inductive load is it?

What are you trying to do?



user1453 said:
what I want to end up with is a current limiting circuit that will work in the 0-300v range and allow regulatate the current to be less than a preset value. The range of this being 1-8A
What's the minimum voltage accross the load?

For example, if it's 150V when the current is 4A a linear regulator is a bad idea since it would need to dissipate 600W.

user1453 said:
there is a H-bridge around the inductor already, I just want to limit the current entering the h-bridge
Is this a motor by any chance?
Are you trying to protect it from overheating when stalled?

A circuit breaker would be a far more suitable option.

Or are you wanting to control the speed?

Because PWM woud be a better choice.
 
As my control circuitry stands at the moment, I have a 555 chip in astable mode outputing a square wave ar roughly 600khz this is then fed into a single transistor inverter to acheive two signals which are half a wavelength out of phase

these two signals are then passed through two seperate totenpole current amplifiers (one for each circuit). This signal (approximately triangular) is then fed into two comparators, these comparators compare the signal to vref. The output of these two comparators are conected to the output of two more comparators these two comparators are running from a signal from a secont 555 timer.

The 4 comparators form two and gate, the output signals for the gate drive are only high if both the input from the interrupter and the pwm sections are high

I included the interupter 555 because I thought it might be possible to use this as an option for reducing heat build up in the power transistors.

As for gate drive that is hadled by two pulse transformers not shown in the digram. Each has two secondarys. the setup is similar to the second digram below

would it be possible to replace the interupter 555 with a circuit based on a hall sensor to limit the current?
 

Attachments

  • finished pwm circuit.JPG
    finished pwm circuit.JPG
    80.3 KB · Views: 1,098
  • AN1schem.gif
    AN1schem.gif
    10.8 KB · Views: 469
600kHz is too high a frequency for a 555. It is driving the dead short of a 10 ohm timing resistor so it will get extremely hot. If it works, its output will be a low level triangle wave.

The Vref is at 0V. Many of the comparators have a low input voltage near 0V so their offset voltage will cause some not to work. I think Vref should be +6V.
 
I was planning to use the cmos version of the 555 they are good up to 1-2 Mhz

Vref is at 6v it is in the middle of a potetial divider, R7 will be increased to about 1k, it is low currently to cut down on simulation time
 
where did you get that gate-drive cct from? - that wont drive at 600kHz with any real power!

Also as I said you are going to need to use IGBT's since you link potential is going to be over 100V

And with IGBT's you are going to need a firmer gate-drive cct.
Likewise you will not be able to switch them at 600kHz, likewise is there any need? 100kHz should be more than enough

How did you come at 600Khz? the normal way is to calc the rise-time of current in yr inductive-load and work out what ripple you can cope with, AND also look at the switching characteristics of yr silicon switch.

You might be missing a trick here...
Do you need bi-directional current? ie +/-8A or ONLY positive amps?
The only reason to use a H-Bridge is because you need current to flow both ways in the load. IF the current is only going to be one-way then a simple single-switch chopper is that is needed

You really got to think abt this one, since yr control cct has not interlock control and you will get soft shoot-through's
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top