EEVblog Electronics Community Forum
General => General Technical Chat => Topic started by: Psi on April 14, 2019, 10:04:21 am
-
I have inherited this charge pump circuit in a product and i'm checking it over to make sure it's all running within rated spec of MAX1044.
The output load draw is 5mA and the output voltage 34V, so 170mW.
Therefore there has to be around 18mA flowing into the system at 9V to equal approx 170mW (excluding inefficiency)
The MAX1044 is obviously pumping C3,5,7 up & down. Adding 9V to the bottom of each stage and lifting up to 18V then 27V and finally 36V.
But does the input current come from the 9V supply direct to D1 or through the MAX1044 or more likely from both? but in what ratio?
The MAX1044 datasheet does simply state a max output current which i'm guessing is because it's not as straightforward as that.
Any thoughts?
MAX1044 datasheet here https://datasheets.maximintegrated.com/en/ds/ICL7660-MAX1044.pdf (https://datasheets.maximintegrated.com/en/ds/ICL7660-MAX1044.pdf)
(https://www.eevblog.com/forum/chat/is-this-max1044-charge-pump-running-within-spec/?action=dlattach;attach=705312;image)
-
It comes from both the 9V and the switch (CAP+), but as the switch node comes from the chips VCC, which is connected to the same supply - does it matter?
Although the output is low current, I would prototype this to double check. Even with low Vf schottkys I think 34V is somewhat optimistic. Granted, the ideal output should be 36V (18V*3) so it leaves 2V for 6 diodes, that doesn't include the inevitable voltage sag from charge pumps. That said, it could be whatever circuit this is in works with that output just fine.
In my limited experience of adding stages to charge pumps, efficiency is still pretty good, but drops sharply when approaching the "advertised" current output. I would hazard a guess at 36v*5mA (this uses ideal output voltage, so will include diode loss), with ~70-80% eff. given the switches are now switching 3 caps instead of 1, + quiescent current. = 180mW/0.7 = 257mW. With 9 V in, 28mA, so say 30mA max.
It would be hard to find an inductor-based boost converter that will take less current than this, its possible for average current, but it will be higher current in bursts.
Edit: LOTS of mistakes.
-
Simulate it.
I'm trying to do a X3 -Ve converter, surprisingly it's not quite as simple as adding a cap and 2 diodes to Cap+, it tends to lock-up instead of starting up.
If I simulate your X4 with a LT1054 the waveforms are below, the last one is the total current into the V+ pin, the second to last is the current in to and out of the Cap+ pin. With 9V it doesn't quite get to 34V@5mA, it manages 32.9V@4.85mA
-
Thanks guys.
The circuit is built and working, i'm just not sure if its working outside the ratings of the MAX1044 because the datasheet isn't clear what the max current the chip can handle on the CAP+ pin
-
Strange the way the data sheet tells nothing about the vital Cap+ pin.
If you're getting 34V @ 5mA from 9.0V, the Cap+ pin's +/- 50mA must be saturating to the rails very well, and dissipating very little, because every 50mV counts. I'd be surprised if the chip's total dissipation is 50mW, so don't worry about it. :)