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.

Step-up DC/DC converter output current

Status
Not open for further replies.
Hi there,


I have tried making a small circuit using the LTC3122 and the application example on page 19 in the datasheet (http://cds.linear.com/docs/en/datasheet/3122fa.pdf), however, with a slightly larger coil. The circuit is supposed to source 833 mA @ 6V, but 70 mA is the max current I can draw from the circuit with a small servo motor or a resistive load. The voltage input is a couple of AA batteries.

How is the current limit controlled by the circuit, and why is the current draw so low before affecting the voltage?


Thanks in advance
 
The DC-current carrying-capability of the inductor in switching converters is CRITICAL. The inductor you substituted is likely saturating, and that prevents the converter from delivering its rated current.
 
First of all - fast answer...

The maximum DC resistance and saturation current is 10.67mΩ/1.5A for the original inductor and 18mΩ/2.74A for the new one.

Is that factor 2 really that important even though the saturation current is way larger for the new inductor? And enough to increase the current output with a factor of 10?
 
Have you measured the battery voltage when the converter is under load?
 
It is interesting in that you are getting 1/10 of what you should. These seems to not be a way of setting the current limit so that's not it.

Mike is right inductor saturation is often the problem.

What voltage is on your batteries when the current is too much? With 3V in and 6V out, the battery current will be 2X the load current. Maybe the batteries can't deliver more than 140mA.

What is the voltage on you PWM/SYNC pin? If the pin is not connected to VCC then this problem will happen.
 
First of all - fast answer...

The maximum DC resistance and saturation current is 10.67mΩ/1.5A for the original inductor and 18mΩ/2.74A for the new one.

Is that factor 2 really that important even though the saturation current is way larger for the new inductor? And enough to increase the current output with a factor of 10?

As long as the new inductor has a saturation current greater than the specified one, then that will not be the root cause of your problem. You didn't give the info about the inductor sat. current in your original post...
 
MikeMl: Sorry about the lack of info in the first post - the second one made up for it

ronsimpson: Yes, the PWM/SYNC pin i 3V and the batteries are indeed capable of delivering more than 140 mA

I have made some measurements with some different resistive loads

Code:
 Load      V_IN     V_OUT    I_LOAD
------------------------------------
OpenL    2.81 V    5.32 V      0 mA
100 Ω    2.74 V    4.73 V     48 mA
50 Ω    2.68 V    4.42 V     89 mA
25 Ω    2.59 V    3.85 V    123 mA
NB: The 70 mA max current was for a specific servo motor and not the upper limit of the circuit.

The battery voltage is affected a bit, but the output is affected a great deal even at low currents (the converter is rated to 2.5A). Also, the 6V is not achieved even in open loop configuration with 1% voltage divider resistors.

Somehow the converter is not allowed to draw a high current without saturating
 
Last edited:
The regulator appears to be not regulating at all.
Have you double checked the wiring and component values?
 
I will check the circuit again, but I don't think that's the issue here
If the circuit is build to a known design with the proper components and wired correctly then it should work, unless one of the parts has failed.

What do you think the issue is?
 
Of course the datasheet circuit is functional, and I have checked the circuit wiring.

Just saying I will try scoping all pins to check for noise or any other wrong signal levels.

But I don't know where to start o_O
 
Hi,

What are you using for Cin? You should try a low ESR electrolytic capacitor. The inductor has to be able to get a huge inrush of current on each switch cycle in a very very short time. If there is any inductance between the circuit and battery, the current could be severely limited which would not allow enough energy to be stored in the inductor.
Likewise if the cap ESR is too high the same thing could happen.

To start, check the current though the inductor (scope). When the inductor is suppose to be charging the current should ramp up pretty straight, and the current should go higher than the output current. When the inductor discharges, it should also be a straight ramp down.
You could also check the duty cycle by looking at the switch output.
We could do some basic calculations and see what the circuit is doing.
If you could post scope pics that would help too.

Also, it would be good if you can list the parts used and link to data sheets or at least port numbers.
 
MrAl : Nice detailed answer :)

I'm using the components recommended by Linear
  • Low ESR caps of the X5R type
  • Low ESR inductor (mentioned in the top posts)
  • Large electrolyt caps and a 100nF ceramic cap across the input voltage pin and GND
I have scoped the VIN pin in different configurations:

No load
LTC3122_NO_LOAD_VIN_VOUT.png


50 Ω load (4.23V and 5MHz)
LTC3122_50OHM_VIN_VOUT_X_MEAS.png
LTC3122_50OHM_VIN_VOUT_Y_MEAS.png


I will perform some additional measurements soon as well as estimating the inductor current.

What duty cycle are you mentioning? No outputs seem to output any logic switch signal with a duty cycle. It will have to be expressed from the inductor voltage then.
 
Hi,

The duty cycle is the duty cycle of the output transistor when it is working at least somewhat normal.
But yikes, is that really the input pin going from some max to zero? That's nasty and surely something is wrong with the feed or the load, or the inductor still doesnt work right.

But it's hard to tell what we are looking at here. Is this the current or voltage?
Also, if the pics where about 2 times larger (wider and higher so 4 times the area) i could see them better. Those small pics are hard to view.
If you use gif files they dont take much space (low byte count) so you can make them a little larger.

Be as detailed as you can with the pics. Tell us exactly what we are looking at. This will help speed the process.
 
Have you made a PCB for the circuit?
The layout is critical, check from the datasheet.

Hi,

Good point. The layout on the high frequency converters is much more critical because the point to point inductance even though small has more of a detrimental effect on the performance. I use mostly lower frequency converters myself (50 to 100kHz) so the layout is not so critical and then sometimes wire point to point on a small perf board. Input cap should be close to the IC anyway and feedback wiring short.
 
jjw : I followed the datasheet guidelines and the entire circuit is made from SMD components taking up approximately ½ sq. inch.

MrAl : Here are the scopes in a larger and readable size (resolution is scope limited)

No load configuration
Yellow: Converter switch pin voltage (SW pin 1), a max of ≈7 V
Blue: Converter output voltage (VOUT pin 11), 5.76 V
It shows a strange repeating pattern of around 1 MHz

scope1.png


50Ω load configuration
Yellow: Converter switch pin voltage (SW pin 1), a max of 6 V
Blue: Converter output voltage (VOUT pin 11), 4.23 V
The oscillation is more stable with a frequency of around 5 MHz (even though it shouldn't be that high, IC has a max of 3 MHz)

scope2.png



I will do some more measurements and maybe make a second circuit with more test points including a proper current measurement functionality. Don't know the time horizon, but I will get back to this thread, if/when something new turns up.

Thanks again


Also,....
...wider and higher so 4 times the area...
Now that's what I call a guideline :)
 
Hi again,

For problems like this it is good to know EVERYTHING about the real life circuit or at least as much as humanly possible.
The pic's are much better now, as i can see the waves much better and start to understand eventually what is going on here.
What is missing now is what values are being used for the real circuit, such as:
Inductor value in say uH,
output cap value in say uF,
input cap value,
lead length between circuit and input power source,
what kind of input power source and what voltage level,
an input voltage level measurement with and without the circuit connected, and with and without load,
the values that determine the switching frequency,
input current with and without load,
an actual ohm meter measurement of the inductor series resistance,
part number for the inductor,
maybe more measurements.

These are measurements and particulars which will help determine what is going on. In a boost circuit everything has to be known or else it's impossible to figure out what is wrong. This is even the case with a buck really though. If you have checked the basics then it is necessary to look closer and a closer look requires more measurements and more information about the real circuit. You can see even the lead length between circuit and input power is good to know, as well as what kind of wire is used to wire it up.
The more information about the circuit that is known the easier it is to find out the unknowns such as what is making that 1MHz wave.

If the layout is 1/2 inch square that sounds very good. So next try to list everything about the circuit you can and make those other measurements.

Also, what is the minimum frequency you can set the chip to run at, is there some minimum?

LATER:
I almost forgot to ask, why do we see the waveforms starting at the middle of the scope shot on the time axis? We need to see the steady state output not the initial startup, which can be very hectic for a boost converter.
 
Last edited:
MrAl : Sorry for the later answer, but I will consider all your information from your past threads in my circuit investigation and later development of a version 2.0

Once again thanks for your interest and knowledge sharing
 
One question that I had was related to the energy storage coil: Is its inductance about the same as that of the original circuit? (If this was mentioned, I managed to miss it...)

In VERY general terms and given a constant frequency, the higher the inductance, the lower the quiescent current but also the lower the maximum current; Conversely the lower the inductance, the higher the available current but the higher the quiescent current.

On one of my early "builds" of a boost-type switching regulator I inadvertently slipped a decimal place when selecting the inductor (e.g. 220uH rather than the intended 22uH or something like that) and was pulling my hair out trying to figure out why it was so "wimpy" - that is, it wouldn't carry a load. It wasn't until after I pulled the inductor and measured it that I'd realized the mistake - one that has caused me to use vigilance (and my inductance meter!) whenever I build these things.

(The selection of the proper diode is also of very high importance as an "ordinary" diode may not work, but I believe that that was covered...)
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top