Max3232 not giving enough V+ ... different chip or different circuit?

Using Arduino Project Guidance
1

I'm making an RS-232 interface for an old serial device, and in the past used a MAX233CPP, and a newer version uses the DIP version of the MAX3232. So far so good.

The problem is that I'm having trouble getting the SOIC version of the MAX3232 playing nicely.

The device I'm using needs a fairly high voltage on its supply. I can tell when it's underpowered because it communicates button presses but not axis motions (it's an old 3d game controller). So the circuit is working in that it is powering switches and communicating correctly to the Arduino, but it's not getting quite enough power.

I think this may be due to the Max3232 itself. I measure 5.3v at pin 2 (V+) and -5.3v at pin 6 (V-) for the SOIC version relative to ground (which makes the device act underpowered; V+ is connected to pins 4-7 on the DB9, with pin 5, ground, connected to ground); the board I have using the DIP version is a little better at 7.4v (device seems happily powered). But the old Max233 is showing a sturdy 10V or so (device wuvs it).

So... does the 3232 simply only put out 5.5v or so? Would I be better served with something like the 232A, or is there some odd way of connecting the 3232 that would do the job better? (I toyed around with connecting V- to the ground pin of the DB9, so the difference it "saw" was 10v, but I'm not sure what that would do to the signal. May try it anyway but it doesn't sound like the "right" way).

2

That's all well and good, but I really don't understand how/what you're using the MAX3232 for from that description.

It sounds like you're using it as a poor-man's DC-DC convertor...?

The current capabilities of the MAX3232 don't lend itself well to doing that.

It would help if you posted some schematics of what you are doing.

3

I'm just using it as a RS232 level converter like it says on the box :), it's just that my RS232 device wants more voltage. I've attached the relevant snip of the datasheet and of my schematic; I'm not doing anything too wacky. I'm connecting Pin 2 (V+) to be the source of voltage on the DB9, but I see that done in a number of example circuits to provide power for an external standalone device (and did it that same way on the last iteration of this device, and it works quite sturdily, good communication, etc).

I've actually seen a number of references saying that the 232a provides +10v and the 3232 provides +5.5v, so I'm suspecting that's my answer and I need to order different chips, but I thought I would at least get other opinions.

max3232.PNG

schematic.png
schematic.png755×414 6.96 KB

4

Well, the internals are basically a voltage doubler.

The MAX232 is a 5V device, so it should be able to pump that voltage up to 2x5 = 10V.

The MAX3232 is a 3.3V device, so it should be able to pump it up to 3.3x2 = 6.6V.

If you're running it at a lower voltage, then it should pump it up to double that voltage.

5

I'm running the 3232 at 5v actually (it seems fine with this, and the datasheet describes acceptable VCC as 3.0 to 5.5v ). But it just doesn't seem to so much be doubling charge.

Actually, I think I may have found the relevant paragraph on the datasheet, finally:

The MAX3222/MAX3232/MAX3237/MAX3241’s internal
power supply consists of a regulated dual charge pump
that provides output voltages of +5.5V (doubling charge
pump) and -5.5V (inverting charge pump), regardless of
the input voltage (VCC) over the 3.0V to 5.5V range. The
charge pumps operate in a discontinuous mode; if the
output voltages are less than 5.5V, the charge pumps
are enabled, and if the output voltages exceed 5.5V, the
charge pumps are disabled

So it sounds like it's not meant to give me more than 5.5v, and the one that's giving me over 7v on my other board (and my testers' boards!) is basically being nice, and I just designed with the wrong part. Durn it. I'll order some 232a chips and see what sort of result I get.