MAX232 Driver causeing me grief

[sodeaf]

Hey Guys,

I do apologize for this isn't a direct Picaxe Question, although I have 2 20x2's on my Board. But I was wondering if you could help me out.

I built a little demo board using 2 x 20x2 and a RS232 Driver. I was using a Linear regulator but unfortunately it was getting to hot and I decided to switch to a Switching regular direct replacement (7805). The new one is a Recom R-78E5.0-.5 500mA. My circuit only draws a maximum of 200 mA.

The problem I am having.. Is I have 2 devices plugged into the MAX232 Driver which are independently powered with there own supply. If I remove 12VDC from my board and leave the devices plugged in to the MAX232 chip I am getting Negative .283 Volts on the output pin of the Switching regulator. The switching regulator will not turn on because of this. When I apply power to board the regulator will not turn on. If i remove the MAX232 connections RX/TX from each device then power board the regulator kicks right in.

Is there a simple way I can fix this issue other then putting a separate small linear regulator at the MAX 232 to isolate it from the system? If I remove one of the Pairs of RX/TX the voltage drops to negative .163 and the regulator will then turn on about 50% of the time. I have checked and re checked the schematic..

Thanks for any input you can give me.. I have googled it all day and cant seem to find anyone with the same issue. I tested a couple older boards I have with different makes and model of MAX232 chips and they all do the same Negative .163 volts . This new board is the first one I have made with 2 x RS232 connections so its doubling the negative voltage causing this.


Steve

 

[rq3]

Steve, it's not clear from your schematic what R20 (10K) is for? At a quick glance referencing the data sheet every thing else looks fine. The only other thought I have is that most switching regulators have to see some minimum load before they will start to work. Recom doesn't specify this, so it may not be necessary, but a 100 ohm resistor from the Recom 5 volt output to ground would give it a 50 mA load regardless of the rest of the circuit. This may give it the kick it needs. You might later increase the resistor to something more reasonable (say 1 K), so you aren't asking the poor regulator to start out at 10% of its total current rating.

Just random thoughts after a quick look!

 

[sodeaf]

Yeah the 10k is a pull up for one of the picaxe pins. It's strange, if I cycle input power to the regulator with the Rx-Tx connected , it won't boot up, as soon as I unplug those connection it works fine .. Frustrating.

 

[sodeaf]

I would think in commercial designs they would have some type of isolation to avoid this . I took a parts couple devices and check to see if the max232 behaves the same way and they do. They all seem generate about -.164 volts of leakage current when one half of the max232 is being used.

Now that I have built a device using both channels it's up to .263which the switching regulator doesn't like at all

 

[inglewoodpete]

You could try inserting a small high-side switch transistor (or FET) to the Vcc pin of the MAX232 configured so that the transistor only turns on when the regulator fires up. That should keep the nasties away from the regulator until it starts. It's not something that I've tried but it might be worth experimenting with.

 

[sodeaf]

Yeah that might work . Use a small FET, I will have to measure current draw of the max232 I can't see it being much.

I am honesty surprised this was even an issue. I've made many projects with the max232 without issue. I am going to try and swap chips just to make sure.

I appreciate the input and will follow up when I figure out the solution.

Steve