0
\$\begingroup\$

I am working on a remote gate opener setup. It will be powered via a 12v battery (12-13.5v) and charged via pwm solar charger at up to 14.5v. I would like to use the HT7333 LDO datasheet here

It has a max input voltage of 12v and I would like to avoid the magic smoke :)

I am looking for a way of reducing the voltage to a safe range so the LDO will work while not draining the 7 Ah battery much when on standby.

Background: The small micro controler will sleep for 1000ms then wake, turn on radio and listen for a signal for a few ms then back to sleep so the bulk of the time the current will be < 20 uA however when awake and opening gates its CPU:10 mA + 2xrelays 70~150 mA.

My thoughts so far:

LDO vs Switching Regulator - LDO as its sleep current is only 4 uA most SR are in the mA range and will empty my battery in winter. Also I have long range radio and the switching of the regulator will cause issues or make my PCB design more complex.

Use another LDO with voltage range i need - There are many designed for greater voltages but nothing in that class comes close to 4 uA current while CPU sleeps in fact most are 2-3 mA.

Use diodes to reduce voltage before the LDO - This seems the most promising as they will only consume power as it is drawn and not much if anything when not. However from the datasheets I have looked at it appears voltage drops on diodes are not static and change significantly based on the current load, so keeping it in a safe range could be tricky and I don't want so many that I have to extend the PCB (could fit in 4)

So I thought I would ask smarter people than I :) thanks in advance for your help.

\$\endgroup\$
4
  • \$\begingroup\$ Borderline XY problem. Don’t use HT7333 and choose an appropriate LDO for the application. \$\endgroup\$
    – winny
    Jul 8 '20 at 23:09
  • \$\begingroup\$ Yes Borderline,that is where I started but low quiescent current and high input voltage is rear it seems especially if you also need affordable. I was hoping someone knew how to reduce the voltage pre LDO \$\endgroup\$
    – AndrewT
    Jul 8 '20 at 23:32
  • \$\begingroup\$ @AndrewT do my suggestions not match your need? \$\endgroup\$
    – BeB00
    Jul 8 '20 at 23:34
  • \$\begingroup\$ @BeB00 sorry no the LDO must have no more that 12v current rating is fine \$\endgroup\$
    – AndrewT
    Jul 8 '20 at 23:36
1
\$\begingroup\$

Use a 3.9 V zener diode. At the 4 uA quiescent current it may only drop about 3 V, but that's still safe for you. At high load (10 mA), it will be very close to 3.9 V.

A lower V zener won't give you as much margin; a higher voltage one will limit you as the battery discharges.

I presume the MCU doesn't (can't) directly drive the relays -- likely you use a transistor (e.g. 2N2222). The coil of the relay doesn't have to connect to the LDO -- it can be connected to the full input.

\$\endgroup\$
3
  • \$\begingroup\$ Interesting, think we are getting somewhere, Yes I am using transisters to drive the coils but they are all part of the 3.3v domain powered by the LDO so max current through it is <200ma \$\endgroup\$
    – AndrewT
    Jul 9 '20 at 0:09
  • \$\begingroup\$ All my experience with Zener Diodes are in parallel where it dumps voltage above a value, how does it work in serious? work it produce a voltage drop at the zener rated value? and like in the forward direction will it still very based on the current? I have a 5v Zener at home is the expected result that if I apply 14.4v ~9.4 will go to the ldo? Sorry this area is new to me \$\endgroup\$
    – AndrewT
    Jul 9 '20 at 4:06
  • \$\begingroup\$ Proven on a bread board, although there is still variability like normal diodes the Zener in reverse bias is far more consistent and should work for what I need. Thanks @jp314 \$\endgroup\$
    – AndrewT
    Jul 12 '20 at 1:44
2
\$\begingroup\$

Zener diode in series with the input could work, if you select one with large enough voltage drop at the rated 4 uA.

Or, change the LDO.

\$\endgroup\$
0
2
\$\begingroup\$

This isn't meant to be a shopping site, but AP7370 and TPS7A25 should do what you want.

In very low duty cycle applications, it's common to use an LDO instead of a buck regulator. You can also use both in parallel, turning on the buck when you need more current.

See here for more info on LDO+buck

\$\endgroup\$
14
  • \$\begingroup\$ Very interesting concept, but not what I am trying to solve, the LDO can handle the full range of current required but even on sleep will explode becuase the input voltage is too high the suggestion above wont solve that unless I have missed somthing. \$\endgroup\$
    – AndrewT
    Jul 8 '20 at 23:35
  • \$\begingroup\$ @AndrewT you said that the input max voltage is 14.5V, is that not correct? \$\endgroup\$
    – BeB00
    Jul 8 '20 at 23:37
  • \$\begingroup\$ yes +/- a few mv relying on a soso solar charing module \$\endgroup\$
    – AndrewT
    Jul 8 '20 at 23:40
  • 1
    \$\begingroup\$ I'd add that I built a 24v to 3v3 switcher with a tps560430 and that consumes<100uA quiescent, and will be more efficient with actual current draw... And the PCB space is basically the same as the lm1117 it replaced... OP really hasn't looked very hard... \$\endgroup\$ Jul 8 '20 at 23:51
  • 1
    \$\begingroup\$ @AndrewT AP7370 is $0.10 in qty 1k, are you saying that the HT7333 is 1 cent? \$\endgroup\$
    – BeB00
    Jul 9 '20 at 1:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.