MAX232CPE pinout different from normal MAX232?

what's a normal MAX 232? They all have a suffix.... and they all have the same pinout. Differences are the type/size of cap you put round them...

On a sunny day (Wed, 1 Jun 2011 16:30:40 +0100) it happened "TTman" wrote in :

That is what I thought, but these get freaking hot that way and cause a foldback in the power supply (I tried 3). Replacing with an older MAX232 shows no problems. These are marked MAXIM MAX232CPE, got them from ebay UK... Looks like a defective batch if it is not a connection issue.

I scoped the voltages, you see normal + on pin 1, and - on pin 6. You also see the voltage converter waveforms. But once you send data to it (from the PC serial port) these things behave like they go into latchup! I use IC sockets, so it is easy to compare the old (good ones) and the new (no good) ones.

Was hoping it was a connection issue. If not, 20 chips for the dustbin. Not a big deal, payed 10 Euro for it IIRC. Will measure some more stuff.

e

oldback

w (no good) ones.

In case you're interested, the intersil HIN202 (16pin DIL) is pin compatible with the MAX232CPE and requires only 0.1uf caps.

TomC

On a sunny day (Wed, 1 Jun 2011 11:02:44 -0700 (PDT)) it happened tomcee wrote in :

OK, I will, if I cannot fix this, look for those too. Thank you.

Check if you got counterfeit parts. Or maybe rebranded Sipex parts (be sure to avoid Sipex parts).

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

Ditto the MAX202(CPE) - we switched years ago, getting away from electrolytics in the process.

I look at the Max205 and don't see how the voltage gets doubled on the outs?

What am i missing? Does it have internal caps?

Jamie

It still has external capacitors -- C1 and C2 form part of a charge pump that doubles and inverts the voltage (see figure 4); I presume they were able to get the capacitors down to 100nF just by increasing the switching frequency.

Did you actually look at the MAX205 lay out (Page 14)?, there is 24 pins., all accounted for. None of them have anything to do with charge pump caps? all of them are inputs/outputs ENable, shutdown, Grd and input (5+) = 24 pins.

I must be blind, because I see no pins available for external charge pump caps..

Jamie

On a sunny day (Wed, 01 Jun 2011 21:58:28 GMT) it happened snipped-for-privacy@puntnl.niks (Nico Coesel) wrote in :

I think I found a clue late last night... I noticed that if I grounded the scope alligator clip on the metal tab of the 7805 regulator, I could start the thing up and connect it to the PC and it would work. Without the scope it would go into latchup and the MAX + and - voltage generator would stop, the MAX232 would get extremely hot, and the 7805 would go into folddown because of the current. BTW my scope is NOT grounded (main lead). So it seems the capacitance from the 7805 ground to 'the rest of the world' makes a difference. Because the supply lines are clean, and decoupled at the pins of the MAX with 10 uF, at the output of the 7805 with 10 uF, and further away with 220 uF(!), I started thinking 'start up problem (risetime)'. But a MAX232 should not care about a slow rising supply, and scope would make things worse, not better. As the RS232 ground is grounded *at the MAX pin*, and the RS232 leads are also connected directly to the MAX pins, that could not induce anything either. BUT further on the board is the dual transistor oscillator that drives the fluxgate core into saturation... Now those pulses are extremely fast, and almost in the 1A region (peak), maybe more at startup, so my theory now is that there is a huge voltage drop on the ground, nano seconds, biggest at start up, that causes the input / output connection of the MAX232 on the PIC side to go NEGATIVE relative relative to the MAX232 ground. This then causes latchup. I will experiment a bit by changing the ground wiring for the fluxgate driver to a separate wire to the 7805, so this situation cannot occur. I cannot scope the effect, my scope is not fast enough, but this is my theory after giving the observations some thought. I will report if that fixes it. Why the older MAX232 chips do not go into latchup, and these do, I dunno. I do not see any sign of rebranding on the chips, but of course they can make labels really well the days.

This is actually the real fun part of electronics :-)

On a sunny day (Wed, 01 Jun 2011 21:39:17 -0400) it happened Jamie wrote in :

My datasheet, page 1: MAX235 (MAX205) +5 5/5 0 -- Yes -- 120 No external caps

Power No. of Nominal SHDN Rx Part Supply RS-232 No. of Cap. Value & Three- Active in Data Rate Number (V) Drivers/Rx Ext. Caps (µF) State SHDN (kbps) Features MAX220 +5 2/2 4 4.7/10 No --

120 Ultra-low-power, industry-standard pinout MAX222 +5 2/2 4 0.1 Yes -- 200 Low-power shutdown MAX223 (MAX213) +5 4/5 4 1.0 (0.1) Yes 120 MAX241 and receivers active in shutdown MAX225 +5 5/5 0 -- Yes 120 Available in SO MAX230 (MAX200) +5 5/0 4 1.0 (0.1) Yes -- 120 5 drivers with shutdown MAX231 (MAX201) +5 and 2/2 2 1.0 (0.1) No -- 120 Standard +5/+12V or battery supplies; +7.5 to +13.2 same functions as MAX232 MAX232 (MAX202) +5 2/2 4 1.0 (0.1) No -- 120 (64) Industry standard MAX232A +5 2/2 4 0.1 No -- 200 Higher slew rate, small caps MAX233 (MAX203) +5 2/2 0 -- No -- 120 No external caps MAX233A +5 2/2 0 -- No -- 200 No external caps, high slew rate MAX234 (MAX204) +5 4/0 4 1.0 (0.1) No -- 120 Replaces 1488 MAX235 (MAX205) +5 5/5 0 -- Yes -- 120 No external caps MAX236 (MAX206) +5 4/3 4 1.0 (0.1) Yes -- 120 Shutdown, three state MAX237 (MAX207) +5 5/3 4 1.0 (0.1) No -- 120 Complements IBM PC serial port MAX238 (MAX208) +5 4/4 4 1.0 (0.1) No -- 120 Replaces 1488 and 1489 MAX239 (MAX209) +5 and 3/5 2 1.0 (0.1) No -- 120 Standard +5/+12V or battery supplies; +7.5 to +13.2 single-package solution for IBM PC serial port MAX240 +5 5/5 4 1.0 Yes -- 120 DIP or flatpack package MAX241 (MAX211) +5 4/5 4 1.0 (0.1) Yes -- 120 Complete IBM PC serial port MAX242 +5 2/2 4 0.1 Yes 200 Separate shutdown and enable MAX243 +5 2/2 4 0.1 No -- 200 Open-line detection simplifies cabling MAX244 +5 8/10 4 1.0 No -- 120 High slew rate MAX245 +5 8/10 0 -- Yes 120 High slew rate, int. caps, two shutdown modes MAX246 +5 8/10 0 -- Yes 120 High slew rate, int. caps, three shutdown modes MAX247 +5 8/9 0 -- Yes 120 High slew rate, int. caps, nine operating modes MAX248 +5 8/8 4 1.0 Yes 120 High slew rate, selective half-chip enables MAX249 +5 6/10 4 1.0 Yes 120 Available in quad flatpack package

ks

nerator would stop,

ecause of the current.

d' makes a difference.

with 10 uF,

started thinking 'start up problem (risetime)'.

make things worse, not better.

also connected directly to the MAX pins,

e fluxgate core into saturation...

rop on the ground, nano seconds, biggest at start up,

o go NEGATIVE relative relative to the MAX232 ground.

iver to a separate wire to the 7805,

eory

make labels really well the days.

Sounds to me like a grounding problem between the power supply for your circuit and the PC itself. If the PC's ground is "floating" you're just asking for troubles.

-- Joe

On a sunny day (Wed, 01 Jun 2011 21:58:28 GMT) it happened snipped-for-privacy@puntnl.niks (Nico Coesel) wrote in :

I will experiment a bit by changing the ground wiring for the fluxgate driver to a separate wire to the 7805, so this situation cannot occur. I will report if that fixes it.

I changed the wiring so the fluxgate stuff has its own supply wires, and took a fresh MAX2322 from the same batch. It has been working now for 10 minutes without problems?

Took a picture of a MAX that died, and you can feel with your fingernail that the

printing is in a slightly lower area than the rest of the chip. That is the grey area, and the chip is dark black (small edge in front. So maybe they are rebranded, do not know for sure. Would MAXIM bother to carve out some of the package? And these are very readable, not normal either :-)

This is fun too, but AMD makes very good chips, its still working.

I got fooled, cleaned the PC inside, started in BIOS health, it displayed 46 C, but I did smell paper.

But it kept steady at 46 C, now I know the BIOS hanged (the CPU halted) Cooled of, mounted correctly, been on now 12 hours or more. Better get a spare Duron.

On a sunny day (Thu, 2 Jun 2011 04:58:06 -0700 (PDT)) it happened "J.A. Legris" wrote in :

Nope, it also does latchup (did) without the PC, and the PC is of course grounded. It is fixed now (I think), see my other reply.

Ah, I'm sorry, no -- I was having a dyslexic moment, it seems, and read your part number as MAX202 rather than '205.

For the MAX205, I would guess that you're correct about their using internal capacitors... although perhaps even markedly smaller than 100nF.

---Joel