1. Home
  2. Electronics Design
  3. Max 232 problem

Max 232 problem

I have a circuit that is misbehaving. The DTR pin on a D9 goes to a MAX

232ACPE input, the output goes via a MAX 707 to control reset and ISP on a Dallas 89C450.There is data in and out from the 232 chip to a PC for programming. The 232 has 100N's for the boost. Common circuit, works 100%,except in this case. Programming the CPU and the system works 100%.... EXCEPT I have noticed occasional resets to the CPU occurring.On further investigation, if I put a piece of insulated 7/0.2 wire with the end stripped and tinned ( 6") into the DTR pin on the D9 socket, I can make the processor reset! Clearly there is a spike being generated by doing this on the DTR line. I can also generate a spike on the data in pin by doing the same thing . The CPU RXD sees a start bit and I see my RXD activity led flash. The system is powered by a mains switching PSU. There are several of these systems and all exhibit the same 'fault'.I have eliminated the max chip by substituting other man. parts and all do the same thing.This is driving me nuts. Can anyone shed some light as to what the problem may be, or where to start looking.? I defies belief as there is a 5K pull down on the Rin inputs..... How can I possible generate a spike when there is this low impedance at the input pin? TIA
Reply to
martin.shoebridge
Loading thread data ...

100%....

the

to

I

pin?

Does your power supply have a mains filter and if so, is it grounded? If not, your circuit floats around half the mains voltage and when you stick that wire in the DB-9, you'll effectively put half the mains voltage on a signal line.

Meindert

Reply to
Meindert Sprang

SNIP

OK That makes sense.... I have powered the system with a wall brick and I can't make it go wrong.( perhaps I should have mentioned that.... Anyway, being an analogue dumbo( sadly) how would I tell if the filter was grounded? I assume there is one fitted as the PSU is CE approved ( or am I being naive?)- It's a switch mode +12V 10 amp thing... Can I measure that half mains? from Mains earth to logic 0V? I'm sure the PSU mains input ground is strapped to the frame of the PSU, as is the 0V on the 12V supply. I'm pretty sure I tried it on a 3A bench supply and it still showed the fault........ Thanks.

Reply to
martin.shoebridge

5k is low impedance ? how2 about having 100nF to GND at the input of the line ?

Rene

--
Ing.Buero R.Tschaggelar - http://www.ibrtses.com
& commercial newsgroups - http://www.talkto.net
*** Free account sponsored by SecureIX.com ***
*** Encrypt your Internet usage with a free VPN account from
http://www.SecureIX.com ***
Reply to
Rene Tschaggelar

Measure between the power supply output and the mains protective ground, with your meter set to AC.

Meindert

Reply to
Meindert Sprang

Shouldn't have to do that...but it would be a fix. Trouble is, that will kill the 56K baud comms. The current fix is to connect a plug with DTR and RXD connected to ground. Bit it's only a work around....

Reply to
martin.shoebridge

The typical model for the human body is several nF. If you walk across a nylon rug you're charged up to several kilovolts. So when you touck that pin you could be discharging a 3nF capacitor charged up to 1KV, easy.

RS-232 inputs have considerable noise immunity, but that's ridiculous.

If you don't want the CPU resetting on DTR line spikes, put a rolloff capacitor from input to ground, or better yet add a 1K resistor before it to help reject noise even even more.

Spikes on the data line you could almost ignore, unless your software could go wacky with random stray bytes coming in.

Reply to
Ancient_Hacker

Which is bad, because you force the signals on a level they should avoid. A valid level on RS-232 is higher than +3V (logical 0) or lower than -3V (logical 1).

Meindert

Reply to
Meindert Sprang

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.