PTN04050c not behaving as expected

[McDude]

I've tried this with two seperate PTN04050c chips and get the exact same result. The circuit is correct and the output is 5.05v consistent. However, when a substantial load is introduced (such as the atomizer) the output of the chip drops to 3.3v (input to the circuit is 3.5v).

So, what I'm wondering is: what the heck is going on? The caps are in the right spot, the devices are functioning as expected (its a very simple circuit) and yet when a load is introduced the voltage drops?

I've scanned thread after thread for mention of this and nada. Given that I've wired-up two different PTN04050c devices (different packaging version is all) with the exact same results, I'm left with two possible cnoclusions:

1. My schematic isn't right (which can't be, it's just too simple and I get the expected output w/out a load)
2. Voltage drop is normal and no-one has bothered to verify? I've read posts about circuits with PTN04050's burning up addy's, so I doubt the voltage drop is normal.
3. The voltage drop is normal and somehow compensated for in an expended schematic (makes no sense)

Anyone with some experience with the chip, please chime in - really stumped by this behavior. If I were to believe in "magic" - I'd say the chip is converting some volts to cover the amp debt when the atomizer is brought into the equation - but then again, I don't ride a unicorn either, so am entirely unsure

 

[AttyPops]

What battery are you using? What ohm atty? Just a thought: since you are trading instantaneous amps (not MAH) for instantaneous volts, you'd need a lot of amps (it will, of course, affected the duration of usable charge from the battery). I've read that 18650's are good. You might want the high-drain stuff. Just a thought. I'm sure others will chime in soon.

 

[CapeCAD]

Voltage drop is normal as load is increased. You should be able adjust the output under load to your desired voltage up to the maximum power rating (12 Watts on this device). The resistance of your atomizer will affect the maximum voltage you can run before you reach this limit.

 

[Nuck]

The battery is the likely cause. Anything below an 18500 is pretty iffy. The 18650 2600 mah Tenergy, the 2600 and the 2900 mah AW's are all good choices. You can also go with the IMR bats if you don't mind reduced cycle time and even then you would require a few 14500.

The only other cause I can think of would be a bad/cold solder joint.

 

[McDude]

Voltage drop is normal as load is increased. You should be able adjust the output under load to your desired voltage up to the maximum power rating (12 Watts on this device). The resistance of your atomizer will affect the maximum voltage you can run before you reach this limit.

That's different behavior than, let's say, a linear 5v regulator where such a voltage drop is not observed. So, basically, under load I should just calculate the volts needed and tune the 4050 for that, like 12 volts or something to get an effective 5? The device spec states 2.5amps (or thereabouts) at 5volts - should be plenty to drive a 3ohm atty without a 2.5 volt drop?!?

 

[McDude]

The battery is the likely cause. Anything below an 18500 is pretty iffy. The 18650 2600 mah Tenergy, the 2600 and the 2900 mah AW's are all good choices. You can also go with the IMR bats if you don't mind reduced cycle time and even then you would require a few 14500.

The only other cause I can think of would be a bad/cold solder joint.

The power is coming from a dedicated "lab power box" that is variable volt regulator and a laptop AC/DC power supply. It routinely supplies 7.2v at 3.5amps easily. For this application I have it dialed to a 6.75v output. Tried the same on batteries, same result, 40% voltage drop.

 

[McDude]

Errr. correction - lab power dialed to 3.5v for this application - been mixing my step-up and step-down circuits today so been dialing it in for different values

 

[Nuck]

The power is coming from a dedicated "lab power box" that is variable volt regulator and a laptop AC/DC power supply. It routinely supplies 7.2v at 3.5amps easily. For this application I have it dialed to a 6.75v output. Tried the same on batteries, same result, 40% voltage drop.

You have the output of your power source set to 6.75? The power source should be set to about 0.5v less than the required output of the booster.

Edit: Missed your new post

 

[Nuck]

Given your target of 3 ohms resistance at 5v you should be fine (about input 2.8amps @85% eff). The odds of 2 boards having issues are pretty small. I've built hundreds of mods with that board and only managed to kill one (my fault). I have seen the problem before and the cause was inevitably a bad solder joint.

 

[CapeCAD]

Something definitely sounds wrong, you should not see this big of a voltage drop with a 3 Ohm atomizer as long as the input can supply the current.

 

[McDude]

Voltage drop is normal as load is increased. You should be able adjust the output under load to your desired voltage up to the maximum power rating (12 Watts on this device). The resistance of your atomizer will affect the maximum voltage you can run before you reach this limit.

I just tested a 5v step down (UCC283T-5) and found about a .5 volt drop due to load (same atomizer) - went from 5.05v to 4.5 on load. The boost circuit (PTN04050c) has a 2.5v drop (same atomizer). /scratching-head

 

[McDude]

Given your target of 3 ohms resistance at 5v you should be fine (about input 2.8amps @85% eff). The odds of 2 boards having issues are pretty small. I've built hundreds of mods with that board and only managed to kill one (my fault). I have seen the problem before and the cause was inevitably a bad solder joint.

One circuit is soldered, one is solderless - same result - completely baffled. voltage dropout on a UCC283T-5 is .5 volt vs. the PTN04050c is 2.5v same atomizer. I must have something not connected in similar ways in both 4050 test circuits.

 

[CapeCAD]

Input current of a boost converter is higher than output current. Input current of a step down converter is less than the output current. I'd also measure your source voltage/current to see if it is supplying the converter correctly under load.

 

[AttyPops]

WAGs:
1) Change your switched-mode power supply feeding the switched-mode booster to a true DC battery supply feeding the booster. This will eliminate the PS being the issue for diagnostic purposes.

2) add capacitors to the input of the booster to smooth out the input and/or verify the caps already on the booster. Maybe you have more ripple in the input current than you think you do.

 

[McDude]

WAGs:
1) Change your switched-mode power supply feeding the switched-mode booster to a true DC battery supply feeding the booster. This will eliminate the PS being the issue for diagnostic purposes.

2) add capacitors to the input of the booster to smooth out the input and/or verify the caps already on the booster. Maybe you have more ripple in the input current than you think you do.

Finally got my Li-ion charger today, so just tested with a freshly charged 14500 and, you were right, using a "true" battery resulted in expected results (boosted 5v and nominal voltage drop - as it should be). Having a hard time understanding why the output from the powersupply would have been the issue, but that's a topic for some research (grabbing my multi-meter and some test leads now! hahaha).

Thanks for the idea. I'm going to test one thing though first, before calling it the PS - I had the PS dialed in for 3.5v where these batteries freshly charged are over 4v - not entirely sure that's not the issue as well - but doubting it.

Thanks a lot.

 

[Java_Az]

I dont think you would have had that problem if you were using a linear PSU. I think this booster needs a direct current to work properly which a linear PSU or battery can supply. I would test it out and see for sure but saddly a linear PSU is on my to get list. Maybe Santa will leave one for me tonight I been much less naughty then previous years.

 

[McDude]

The "lab power" is basically a box containing a 30vdc, 4.5amp laptop power supply that feeds this:
DC-DC IN 3-30V OUT 1.3V-18V converter Module board Z - eBay (item 150538704983 end time Dec-25-10 09:24:15 PST)

I assumed because I'm regulating the output using that board that it would not be an issue - I was definately wrong.

 

[Java_Az]

That board looks like it is switched. Which just means the current is not solid. Depending on the frequency it is turning on and off hundreds if not thousands of times a second. Why it is bad is the booster chip is also switching to boost the voltage. So i think they are conflicting with each other. Might be something about it in the data sheet. I dont think it would have been as bad if supply was on a really low switch frequency. But i will check back in if there is something in the data sheet about it.

Edit: From the data sheet i believe it has to do with noise from the switching DC supply on the frequency the booster uses for switching. Switching regulators are more efficient but do put out alot more noise into the circuit then a linear reg or battery would do. I believe you could get the switching reg to work if you went with a more Noise and ripple reducing set up like this one in the data sheet. But since your going to be running this off batteries i dont think it is needed but might help in understanding a bit better