Maker Pro
Maker Pro

DC Booster IC

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
DC Booster IC - Help Needed

I'm trying to make use of the LT1303 IC to boost 5v to 9v but I've come into a couple of problems, the first is that unless I add a small resistor to Vin the chip gets very hot and the ouput voltage drops. The second is that the output voltage drops depending on the load.

The goal is to power 2 2.3" 7-segment displays from 4 aa batteries, but as more segments come on the volage from the IC drops and the displays dim (or potentially go off).

Does anybody know of something i've missed, I've not done a lot of electronics stuff some I'm not entitrely certain.

I'm following the circuit at the top of Pg 9 from the datasheet ( http://cds.linear.com/docs/Datasheet/lt1303.pdf) with R2 at 120k and R1 at 18k which should give me around 9.5v but with a 10 ohm resitor on Vin I get 8.5v and as more segments on the display are lit the voltage drops further.
 
Last edited:

bountyhunter

Mar 9, 2012
11
Joined
Mar 9, 2012
Messages
11
Do you have a good switching quality cap near the input pin of the IC? That's critically important. Increase it's size and see if it helps.

Also: what is the current capability source powering it? A boost tries to draw very high peak currents, if the source can't supply it, it will not work right.

The symptoms sound like the source is collapsing under load.
 
Last edited:

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
The circuit is currently being powered by a usb emergency charger which gives my 5v at up to 2.1A but it's currently only drawing 175 - 200mA
There's a 100uF capacitor at Vin that I've doubled up to 200 uF with no difference.

With no segments of the display lit I get 8.5v out drawing 175mA at the battery, lighting up one segment puts us at 198mA but the voltage drops to 7.2v lighting further segments has less of an impact to the voltage but it still drops.
 

bountyhunter

Mar 9, 2012
11
Joined
Mar 9, 2012
Messages
11
You need to measure the input currents, DC and peaks. It may be the peaks are high enough to trip the 5V supply's current limit.

You could also have too little inductance or an inductor whose current rating is too low which would cause it to saturate, causing VERY high peak currents.
 

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
I'm beginning to think I've got a faulty chip, with the shutdown pin high the chip still draws 150mA.

Bypassing Vin and only powering SW through the inductor should draw nothing as Q1 in Figure 2 of the datasheet shouldn't let anything through, instead it just acts as a small resistor letting 500mA through it and straight to ground which I'm guessing is what's making the chip heat up so much when I don't add a resistor myself.

There's also always a voltage difference across the inductor which (as far as I can tell) shows that it's not saturated, and using a larger inductor makes no difference either.
 
Last edited:

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
How are you connecting the 7-segment displays? Are you sure they want 9V? Do you have a link to the part?

Bob
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Yep, they really do need about 9V to work. At 9.5V you are going to want 50 Ohms to limit the current to 30ma. If all 14 segments are active this would be 0.42A which given the the boost from 5 to 9V at say 75% efficiency would require about 1A from the batteries. This is really pushing it for AA batteries. The 10 Ohm resistor from battery to regulator would be dropping 10V at 1A, so it is no wonder the voltage drops, I suspect you are getting no more than 100ma out of the regulator with this setup. Basically, your batteries are probably too weak and your are pushing the limits of the chip.

Bob
 

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
At the moment I'm not using AA batteries to power it but a larger battery used for powering USB devices, and it has trouble lighting two segments, never mind all 14
 

BobK

Jan 5, 2010
7,682
Joined
Jan 5, 2010
Messages
7,682
Have you taken out the 10 Ohm resistor? At 5V that is going to drop the voltage by half at 250ma which will get you only about 100 ma out at 9V.

Edited: more like 50ma, I forgot that the input voltage was going to be 2.5V after dropping in half.

Bob
 
Last edited:

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
I can't test anything without the resistor, even with no load it pulls > 500mA which slowly drops off as the chip heats, at the same time the output voltage also drops.
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
OK so there must be a wiring fault

show us the exact circuit you are using
show us a photo or 2 of the project, sharp and well lit

Dave
 

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
Well, here are some pictures of the circuit sans resistor on Vin, the white wire with red tape is Vout.

Circuit_ViewA.jpg

Circuit_ViewB.jpg

Circuit_ViewC.jpg
 

bountyhunter

Mar 9, 2012
11
Joined
Mar 9, 2012
Messages
11
Using a proto board is pretty much a certain way of getting a switcher not to work. The peak currents are too high and ground loop inductance will kill you. Peak current for this part is 1.25A. Having long leads on the switcher caps also seriously degrades their performance because of lead inductance.

I can't see the labels on the big caps, but this is a 150 kHz switcher: are those switcher grade caps with ESR rated at 100 kHz or higher? If those are generic aluminum caps they will never work.
 
Last edited:

AMushrow

Nov 3, 2012
8
Joined
Nov 3, 2012
Messages
8
Hmm, so you think the board could be the cause of my issues?
Any other limitations of a breadboard I should be aware of?
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
Any other limitations of a breadboard I should be aware of?

Yes .... never use them for RF projects, transmitters or receivers :)

Do you have a circuit diag for your construction ?
or did you follow the datasheet exactly ?

Dave
 

bountyhunter

Mar 9, 2012
11
Joined
Mar 9, 2012
Messages
11
Hmm, so you think the board could be the cause of my issues?
Any other limitations of a breadboard I should be aware of?
I spent 20 years as an app engineer at nat semi, and I never saw one of our switchers work right on a proto board. usually like yours: sort of functional but all kinds of strange problems.
 

davenn

Moderator
Sep 5, 2009
14,254
Joined
Sep 5, 2009
Messages
14,254
OK I have downloaded the datasheet

what is the diode you have in there ? I hope its not a 1N400x series one ??
it should be a 1N5817 hi speed switching diode



Dave
 
Last edited:
Top