Dramatic voltage drop at voltage regulator output

[jim380]

First time posting a problem on here. Be gentle on me :)

So I ran into this problem when I was testing the power supply block for my senior project.I attached a pic (https://7b6aa69e-a-59896e8d-s-sites...xn8zXySvNTzHjn4QovXQMIMjf3Sfs=&attredirects=0)showing what my design looks like, and specs that need to be met at each output. The problem was that as soon as I hooked up the servo motor to the entire circuit, the voltage at the green node (output of the 6v voltage regulator) immediately got dropped to around 2 volts. Also, the voltage at blue node as well as the one at purple node got dropped, too. Down below is what my testing procedure looks like:

1. Set up the circuit and everything. (12v battery*1, 5v voltage regulator *1, 6v voltage regulator *1, and a couple push-to-on switches)
2. Hook up display to the circuit. ----Everything Works fine
3. Add micro-controller into the circuit ---- Still fine
3. Hook up servo motor to the circuit -----BOMB! Problem happened

I have been stuck at this problem for a while. Any help would be greatly appreciated. Thanks in advance.


Datasheets:
5v voltage regulator: https://docs.google.com/viewer?a=v&...HV8ZWNlNDR4MjAxMzM4fGd4Ojc2NjQxNzJlOGU1ZDEwZQ
6v voltage regulator: https://docs.google.com/viewer?a=v&...V8ZWNlNDR4MjAxMzM4fGd4OjcxODQ5YzRiNDliNDRlNDU
Micro-controller: ATMEGA32U4
Servo Motor: https://docs.google.com/viewer?a=v&...V8ZWNlNDR4MjAxMzM4fGd4OjQ2ZDA4ZGNjMzM2NzYyYTY

 

[analogdesign]

It looks like the issue is the 6V regulator is shutting down. I presume that if you hook up the servo motor before the display or ucontroller is fails, correct?

It could be any number of things but I suspect the 6V regulator is getting too hot since the servo motor is pulling half an amp. Have you installed a heat sink for the 6V regulator? Does the circuit work when the servo motor is idling (not engaged?)

 

[Baluncore]

Welcome to PF.
1. The regulators need a ceramic capacitor from ground to input and from ground to output. See the test and example circuits in the data sheets. Without capacitors the circuits will oscillate.

2. You need to heat sink your regulators. Do they get hot?

3. Is the 12V battery healthy?
Measure the battery voltage when load is connected, if it falls under load it explains your observations.

 

[jim380]

It looks like the issue is the 6V regulator is shutting down. I presume that if you hook up the servo motor before the display or ucontroller is fails, correct?

It could be any number of things but I suspect the 6V regulator is getting too hot since the servo motor is pulling half an amp. Have you installed a heat sink for the 6V regulator? Does the circuit work when the servo motor is idling (not engaged?)

first of all, thanks for your quick reply.
yes. the 6v voltage regulator is thermal protected, so it shuts down when it's getting too hot. the weird thing is, even if I only hook up a 6v voltage regulator and the servo motor with the battery, the same problem still happens. the voltage regulator gets very hot, and it shuts down by itself.

 

[mjhilger]

My first thought is that the servo circuit/connection is drawing too much current. What happens if you connect the servo power up by itself? Have you put a resistive load on the 6V output to verify it will provide the current you need? If you want to check the 6V by itself 6v/400mA = 15ohms, so load it with a 15 ohm resistor. Or can you grab a lab power supply with current limiting and bypass the 6v regulator you are using?

Try to isolate the problem, but my guess is that the main power exceeds your available current and it is limiting. Also when you connect the servo, it is usual for the motor to seek for a few milliseconds, so you should hear it start to turn briefly even if the power pulls back.

 

[mjhilger]

the 6v voltage regulator is thermal protected, so it shuts down when it's getting too hot. the weird thing is, even if I only hook up a 6v voltage regulator and the servo motor with the battery, the same problem still happens. the voltage regulator gets very hot, and it shuts down by itself.

Have you tested the servo elsewhere? Do you have a different one? Are you sure your connection is not shorting out the voltage? Or maybe you have the + & - of the servo reversed.

 

[jim380]

Welcome to PF.
1. The regulators need a ceramic capacitor from ground to input and from ground to output. See the test and example circuits in the data sheets. Without capacitors the circuits will oscillate.

2. You need to heat sink your regulators. Do they get hot?

3. Is the 12V battery healthy?
Measure the battery voltage when load is connected, if it falls under load it explains your observations.

1. I thought caps are not necessary in my case. Also, I think the circuit example shown in the datasheet is just for demonstration purposes. I'll try adding caps anyways.

2. They indeed got really hot, especially the 6v one, which is connected to servo motor.

3. good point. I will report back once I get to measure it.

 

[jim380]

My first thought is that the servo circuit/connection is drawing too much current. What happens if you connect the servo power up by itself? Have you put a resistive load on the 6V output to verify it will provide the current you need? If you want to check the 6V by itself 6v/400mA = 15ohms, so load it with a 15 ohm resistor. Or can you grab a lab power supply with current limiting and bypass the 6v regulator you are using?

Try to isolate the problem, but my guess is that the main power exceeds your available current and it is limiting. Also when you connect the servo, it is usual for the motor to seek for a few milliseconds, so you should hear it start to turn briefly even if the power pulls back.

Exactly what I thought when I first ran into this problem. A couple things I've tried:
1. Directly power up the servo motor using 4 AA batteries----> Motor works fine.

2. I believe I've tried putting resistive loads (ranging from 2-10 ohms) on the 6v output. But it's like a couple weeks ago, so I don't recall any specific data at this moment. What I do remember is that the voltage regulator shut down itself at some point so the current measured was almost unnoticeable.

3. I do have access to a power supply with current limiting setting. And I already tried that. What I did was that I set the current limit at 0.6A first, and then tried monitoring the output voltage at the 6v regulator as I was slowly increasing the current. It turned out that the output voltage got dropped dramatically when the current is around 0.5A. I assume the 6v regulator just shuts down itself when the current exceeds 0.5A.

 

[jim hardy]

Baluncore and analogdesign have figured it out.

In that datasheet you linked take a look at "Table 2 Thermal Data" on page 2.

How much power is your regulator trying to handle? If it drops from 12V to 6V which is 6 volts, X a half amp, that's 3 watts.

Table 2 says that at 3 watts, even the hefty TO220 package will heat up by 150 degrees C above room temperature. That puts it above its operating range T_op in Table 1 just above.

I assume the 6v regulator just shuts down itself when the current exceeds 0.5A.

It is certainly entitled to. A good heatsink should help a lot.

 

[Baluncore]

3. I do have access to a power supply with current limiting setting. And I already tried that. What I did was that I set the current limit at 0.6A first, and then tried monitoring the output voltage at the 6v regulator as I was slowly increasing the current. It turned out that the output voltage got dropped dramatically when the current is around 0.5A. I assume the 6v regulator just shuts down itself when the current exceeds 0.5A.

I suspect you set the power supply voltage to 12V and the current limit on the high side. Then as you gradually reduce the current limit you will reach the point where the current required is equal to the current limit and the power supply voltage will fall very rapidly in order to limit the current to the lower value set. Once the voltage falls below the drop out voltage of the 6V regulator, (about 8V), the regulator will begin to fail and so it's output voltage will fall.

Your regulator is a current sink, the power supply is a current limited source. In effect the currents are being compared, the voltage on the 12V output will change very rapidly depending on their relative magnitude.

 

[jim380]

It looks like the issue is the 6V regulator is shutting down. I presume that if you hook up the servo motor before the display or ucontroller is fails, correct?

It could be any number of things but I suspect the 6V regulator is getting too hot since the servo motor is pulling half an amp. Have you installed a heat sink for the 6V regulator? Does the circuit work when the servo motor is idling (not engaged?)

Sorry for putting you back on this question, but I'am still suffering from the voltage drop issue. I already tried installing a heat sink on the 6V regulator. When it's unloaded, the output voltage is almost exactly 6V. But when I put a load (a 12 ohms resistor, which is supposed to be equivalent to the resistance of the servo motor I use) onto it, the output voltage drops to 1.7V-ish, which also makes the output current extremely small. Here is how I tested it:

A 12V battery is hooked up to the 6V voltage regulator, and then a load resistor is placed between the output of the regulator and common ground. That's it!

I wonder what is causing this problem. Please help me out. Any suggestion will be greatly appreciated. Thanks in advance.

 

[jim380]

Baluncore and analogdesign have figured it out.

In that datasheet you linked take a look at "Table 2 Thermal Data" on page 2.

How much power is your regulator trying to handle? If it drops from 12V to 6V which is 6 volts, X a half amp, that's 3 watts.

Table 2 says that at 3 watts, even the hefty TO220 package will heat up by 150 degrees C above room temperature. That puts it above its operating range T_op in Table 1 just above.



It is certainly entitled to. A good heatsink should help a lot.

Sorry for putting you back on this question, but I'am still suffering from the voltage drop issue. I already tried installing a heat sink on the 6V regulator. When it's unloaded, the output voltage is almost exactly 6V. But when I put a load (a 12 ohms resistor, which is supposed to be equivalent to the resistance of the servo motor I use) onto it, the output voltage drops to 1.7V-ish, which also makes the output current extremely small. Here is how I tested it:

A 12V battery is hooked up to the 6V voltage regulator, and then a load resistor is placed between the output of the regulator and common ground. That's it!

I wonder what is causing this problem. Please help me out. Any suggestion will be greatly appreciated. Thanks in advance.

 

[jim380]

Have you tested the servo elsewhere? Do you have a different one? Are you sure your connection is not shorting out the voltage? Or maybe you have the + & - of the servo reversed.

Sorry for putting you back on this question, but I'am still suffering from the voltage drop issue. I already tried installing a heat sink on the 6V regulator. When it's unloaded, the output voltage is almost exactly 6V. But when I put a load (a 12 ohms resistor, which is supposed to be equivalent to the resistance of the servo motor I use) onto it, the output voltage drops to 1.7V-ish, which also makes the output current extremely small. Here is how I tested it:

A 12V battery is hooked up to the 6V voltage regulator, and then a load resistor is placed between the output of the regulator and common ground. That's it!

I wonder what is causing this problem. Please help me out. Any suggestion will be greatly appreciated. Thanks in advance.

 

[jim380]

Welcome to PF.
1. The regulators need a ceramic capacitor from ground to input and from ground to output. See the test and example circuits in the data sheets. Without capacitors the circuits will oscillate.

2. You need to heat sink your regulators. Do they get hot?

3. Is the 12V battery healthy?
Measure the battery voltage when load is connected, if it falls under load it explains your observations.

Sorry for putting you back on this question, but I'am still suffering from the voltage drop issue. I already tried installing a heat sink on the 6V regulator. When it's unloaded, the output voltage is almost exactly 6V. But when I put a load (a 12 ohms resistor, which is supposed to be equivalent to the resistance of the servo motor I use) onto it, the output voltage drops to 1.7V-ish, which also makes the output current extremely small. Here is how I tested it:

A 12V battery is hooked up to the 6V voltage regulator, and then a load resistor is placed between the output of the regulator and common ground. That's it!

I wonder what is causing this problem. Please help me out. Any suggestion will be greatly appreciated. Thanks in advance.

 

[Baluncore]

What does the battery and regulator input voltage do when you connect the 12 ohm resistor to the output of the regulator?

 

[jim380]

What does the battery and regulator input voltage do when you connect the 12 ohm resistor to the output of the regulator?

They were around 12V, as they should be.

 

[Baluncore]

If the battery voltage does not also fall drastically then either the regulator circuit is incorrect or the regulator device is faulty.

 

[jim380]

If the battery voltage does not also fall drastically then either the regulator circuit is incorrect or the regulator device is faulty.

what if the battery voltage (or the input voltage at the regulator) does drop drastically?

 

[Baluncore]

If it falls then the battery is dead, probably an open circuit connector or a dry cell.

Are you are definitely using a 7806 regulator chip?
Have you checked your pin numbers? Input, ground and output.

Turn off the power switch on the B+ lead.
Measure the resistance between the common ground and the load resistor ground.
Measure the resistance between the common ground and the battery ground terminal.

 

[jim380]

If it falls then the battery is dead, probably an open circuit connector or a dry cell.

Are you are definitely using a 7806 regulator chip?
Have you checked your pin numbers? Input, ground and output.

Turn off the power switch on the B+ lead.
Measure the resistance between the common ground and the load resistor ground.
Measure the resistance between the common ground and the battery ground terminal.

Thanks. I'll let you know as soon as I get to check it.

 

[jim hardy]

How big is your heatsink?
Can you touch the regulator when loaded or does it get too hot?And yes, you must know whether your battery holds input to regulator around 8 to 12 volts loaded, 11 to 12 would be better.
The regulator can't replace volts that the battery fails to deliver.

 

[jim380]

How big is your heatsink?
Can you touch the regulator when loaded or does it get too hot?


And yes, you must know whether your battery holds input to regulator around 8 to 12 volts loaded, 11 to 12 would be better.
The regulator can't replace volts that the battery fails to deliver.

It doesn't get hot at all. I replace the battery frequently, so it shouldn't be a problem (I guess...)

 

[jim hardy]

okay which 6v regulator do you have?

Google has decided i now need a password to see that datasheet - the same one that worked fine just a couple days ago...
never trust a computer with anything , important or not...

i think i recall that datasheet showing several different packages, not all of which are capable of an amp

Baluncore is on the right track with circuit checks.

 

[jim380]

okay which 6v regulator do you have?

Google has decided i now need a password to see that datasheet - the same one that worked fine just a couple days ago...
never trust a computer with anything , important or not...

i think i recall that datasheet showing several different packages, not all of which are capable of an amp

Baluncore is on the right track with circuit checks.

MC7806B is the one I use: https://www.dropbox.com/s/7bqqb3j565m16bc/MC7806B.PDF

 

[jim hardy]

It doesn't get hot at all. I replace the battery frequently, so it shouldn't be a problem (I guess...)

yet on previous page you said it got really hot...

 

[jim380]

yet on previous page you said it got really hot...

yes. But when I was testing it today in the lab, it didn't seem to get hot, at all. Weird...

 

[Baluncore]

It is only weird because you are overlooking something very simple.
You must use a methodical approach to identify that problem.

Check all voltages relative to one reference point.
Check all joints are soldered and electrically continuous.
Shake the circuit while watching the output voltage to find the broken wire.

When output voltage falls under load...
1.7 volt / 12 ohm = 142 mA.
12V – 1.7V = 10.3V across the linear regulator.
10.3V * 0.142A = 1.46 watt power dissipated in the regulator so it should not get hot.
But why is it limiting the current to 142 mA ?

 

[jim380]

It is only weird because you are overlooking something very simple.
You must use a methodical approach to identify that problem.

Check all voltages relative to one reference point.
Check all joints are soldered and electrically continuous.
Shake the circuit while watching the output voltage to find the broken wire.

When output voltage falls under load...
1.7 volt / 12 ohm = 142 mA.
12V – 1.7V = 10.3V across the linear regulator.
10.3V * 0.142A = 1.46 watt power dissipated in the regulator so it should not get hot.
But why is it limiting the current to 142 mA ?

True. I will try to isolate the problem, and report back. Thanks for your help so far.

 

[jim380]

If the battery voltage does not also fall drastically then either the regulator circuit is incorrect or the regulator device is faulty.

okay. I think I found the problem. When the regulator is loaded with a 12 ohms resistor, the battery voltage drops from 12V to 3.4V . I've tried a couple new batteries, still the same problem.

 

[jim hardy]

Baluncore and jim380 :: ++++++

Jim - speaking of methodical
i like to measure all voltages "with respect to" some common point in the circuit. "With Respect To" just means hook you meter's negative lead to one spot, leave it there and write on schematic all your other voltages.

I'd have tried, in your circuit, to get negative meter lead on negative side of battery and leave it there... on battery itself not just nearby.
that-a-way you'll find the real sneaky troubles like a wire that's broke inside the insulation but looks fine from outside, or a corroded battery terminal like you find so often in automotive troubleshooting.

Methodical-ness is a good habit to develop.

 

[Baluncore]

okay. I think I found the problem. When the regulator is loaded with a 12 ohms resistor, the battery voltage drops from 12V to 3.4V . I've tried a couple new batteries, still the same problem.

Well done.
What type of 12 volt battery are you using, make, model, chemical technology and capacity?

 

[jim380]

Well done.
What type of 12 volt battery are you using, make, model, chemical technology and capacity?

http://www.digikey.com/product-sear...t=0&page=1&quantity=0&ptm=0&fid=0&pageSize=25

I'm using this one.

 

[Baluncore]

Data shows discharge Rate = 480µA
It is doing very well to produce 150 mA.

 

[jim380]

Data shows discharge Rate = 480µA
It is doing very well to produce 150 mA.

Yes. I figured. I also tried powering the 6v regulator using a power generator in my lab. Everything works fine when loaded with the same resistor. I'm literally running out of ideas.

 

[Baluncore]

If you have a lab supply that works, how come you are running out of ideas?
You need to get either an 8V to 12V power supply rated at 1 amp output that runs from the AC mains,
or a 12V gel cell with sufficient capacity to produce 1 amp.