| Author |
Message |
sreed
Joined: 11 Oct 2005
Posts: 13
|
 11 Oct 2005 15:05 load schematic |
|
|
|
|
I repair power supplies on a part time basis. I have been surfing the net looking for an electornic load schematic and managed to only find one which I thought was relatively easy to bulid. I have attached the hand drawn schematic I downloaded. According to the website, big N-Mosfets were supposed to be used but I have used the smaller ones (10pcs of IRF540) and eliminated the sense resistor (1mΩ).The problem is 5 mosfets are running very hot to the point of breakdown and the remaining 5 are somewhat warm. I connected a DC ammeter to the power supply and it only shows a maximum of 4.5A. I would be very grateful if any of you have an idea as to what is wrong with the circuit and how to remedy it. Thanks.
|
|
| Back to top |
|
 |
Desert Leo
Joined: 27 Jan 2005
Posts: 10
Helped: 1
Location: Bulgaria
|
12 Oct 2005 16:35 ac electronic load schematic |
|
|
|
|
Hi sreed!
Your electronic load is a classical current sink circuit or Op Amp based voltage to current converter. Everything is seems to be O.K. , but I am wandering of too small resistance of source resistors - 0.1 ohm.
To distribute load current you need to have resistance that gives you voltage drop (for working current) close to the value of gate to source voltage. That will activate the negative feedback that causes load current distribution. Try to increase the resistance to 0.5 - 1 ohm.
I am afraid that will be hard if you use 10 transistors. Try with less number, but more powerfull ones.
Several years ago, I used same schematics (connected to DC shoulder of bridge rectifier) for a load of small AC generator. So, the schematics is O.K.
regards
|
|
| Back to top |
|
|
Google
AdSense
|
12 Oct 2005 16:35 Ads |
|
|
|
|
|
|
| Back to top |
|
|
VVV
Joined: 26 Nov 2004
Posts: 1584
Helped: 290
|
14 Oct 2005 4:16 electronic load schematics |
|
|
|
|
I think you have two problems:
-load current too low
-transistors not sharing the current equally.
The two could be related.
However, I do not understand how you eliminated the 1mΩ resistors, since it provides the feedback for your circuit. Without it, you will not be able to regulate the current.
Thus, the current being too low could be related to your connecting somehow the feedback of the opamp in such a way that it does sense some parasitic voltage across a trace that acts as the 1mΩ resistor. So check carefully the circuit/ layout.
Another reason could be too low voltage. What voltage is the load supposed to work at? The 0.1Ω resistors will drop some voltage, the MOSFETs too, then there is the wiring and the ammeter. If you are trying to use it at 1.8V or less, chances are you do not have enough voltage to force the current.
The fact that the MOSFETs do not share the current equally could mean the layout is not really symmetrical. Also, the number of MOSFETs seems rather large.
|
|
| Back to top |
|
|
sreed
Joined: 11 Oct 2005
Posts: 13
|
14 Oct 2005 14:34 electronic load transistors |
|
|
|
|
| Thank you very for your input guys.I will try your suggestions and see it will put things right.I hope you won't mind if I come back with more questions.
|
|
| Back to top |
|
|
sreed
Joined: 11 Oct 2005
Posts: 13
|
15 Oct 2005 12:12 op amp current sink load schematic |
|
|
|
|
I've looked at the circiut and I'm sure I have made a mistake. I've need to clarify a few points.Firstly the power supply I'm testing has a 12.5Vdc/6A output.Voltage drop across the 0.1R measures an average of 3.5V at the 4.5A but it does not seem to get hot while the circuit is operating.So do you still think increasing the value will help?
As for the sense resistor, I thought the only function was to provide current measurement and that is why I first try connecting the ammeter at that point but I could only get about 2.5A(analog meter) but with a DMM it measured 0.25A which got me confused. Later I shifted the ammeter to the source point and now both the ammeters gives me the same readings.Is there anything wrong with the way the ammeter is connected and is it giving the actual current drawn by the load? I'm using 10 MOSFETs because that is what the website suggested to provide a maximum of 75A (which I don't really need). So would it be OK for me to use say about 5 big ones?
I don't quite understand what you meant by unsymmetrical layout? Are you saying that there is a certain way the components should be placed on the board or does it have to do with the component values? Thanks
|
|
| Back to top |
|
|
Ante
Joined: 01 May 2004
Posts: 177
Helped: 4
|
15 Oct 2005 14:07 Re: Electronic load |
|
|
|
|
| You should try to make the length of the wiring between point D and point S equal for each MOSFET and resistor. Also the wire gauge is very important for this part of the circuit. I don’t think you need to change the value of the resistors, just keep the 0.1R as is.
|
|
| Back to top |
|
|
VVV
Joined: 26 Nov 2004
Posts: 1584
Helped: 290
|
15 Oct 2005 17:11 Re: Electronic load |
|
|
|
|
So you need to dissipate 75W.
You can use fewer MOSFETs, but with good heatsinks or forced-air cooling.
The layout should be such that the resistances seen by the sources of the MOSFETs are as close as possible. The 0.1Ω resistors should help, but their tolerances will probably be larger than those given by a layout.
If you do not need the 75A, use fewer MOSFETs. That will make things easier on the layout.
I built a similar load that I use in the lab.
One suggestion I have for you is to implement an undervoltage circuit, simply a comparator (use the other half of the opamp) which forces the current to zero if the power supply voltage is lower than a certain value.
This will help you with one thing: when you turn on the power supply, it will turn on into a high-impedance load, instead of a near-short-circuit. You see, when there is very low voltage, the preset current cannot be sunk so the opamp will drive the MOSFETs to saturation. Thus, when you power up the power supply it will see just the 0.1Ω resistors (and MOSFET on resistances) in parallel. That will be only about 0.015Ω. So you turn on the power supply thinking you set the current to 5A, but the load is shorted. The power supply may go into current limit, and sometimes just go into a hiccup mode, not being able to power up at all.
Here is one way of adding that undervoltage, with a threshold of about 2.5V, allowing you to use it down to 3.3V. If you need the load for lower voltages, change the resistors in the divider.
If you need to change the hysteresis, change R4.
|
|
| Back to top |
|
|
artem
Joined: 22 May 2003
Posts: 1652
Helped: 91
Location: Turan
|
16 Oct 2005 0:51 Electronic load |
|
|
|
|
My opinion is that your MOSFET are not identical - that is the reason you cant get current distribution proportionally between them . Try to select them at least with same Vgs cutoff voltage.
because Rds on Vgs characteristics are not same for all transistors and if max load per 1 mosfet will be 10 A - you get 1 V on 0.1 Om connected to source , but actual manufactoring Vgs/Rds characteristics are much wider .
Try to replace them with bipolar transistors (they are usually used in this type of current load) - you will soilve the problem , but darlington approach must be taken as opamp can not supply nearly 1 A source .
Also , may be PWM based approach will be simpler ?
|
|
| Back to top |
|
|
