cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Anonymous
Not applicable
6,255 Views

Update: Solved (fairly certain)....XC9500/XL driving a XR-2203 (similar to ULN2003)

Jump to solution

Update: I did some more testing since posting. It seems I may have had a bad XC9536 to begin with. The 2nd one I used I hadn't tried the non-tristate version of the outputs. I reprogrammed it and have been testing for a while. Seems warm to the touch, but no where near as hot as the 1st one I tried.

 

Glad I figured it out. Did not make much sense given the data I had on the devices.

 

  Cheers.....

 

I have a circuit in which I have an XC9536 (Vcio=5VDC) driving 4 inputs of an XR-2203 (basically a x7 darlington xstor driver to a relay). The XR-2203 specs say it's designed to be driven by 3-5V signal (CMOS/TTL....2.7K in series to the base). I programmed the XC9536 outputs as "normal" and it got very hot (may have even burned out 1 chip). I programmed another XC9536 setting outputs as tri-state (OBUFT) and using a pull-up to 5V. The CPLD still gets fairly warm.

 

Is there any special consideration when using a CPLD like this to drive xstors?

 

  Thanks in advance.

 

   Cheers...Steph

0 Kudos
1 Solution

Accepted Solutions
Highlighted
Professor
Professor
7,408 Views
Registered: ‎08-14-2007

Re: Update: Solved (fairly certain)....XC9500/XL driving a XR-2203 (similar to ULN2003)

Jump to solution

Additional inductance won't help.  Other than capacitance, your best bet is to add another clamp diode directly across the relay coil rather than relying only on the internal diodes of the XR-2203.  That will keep any current spikes local to the relay rather than injecting them into the rest of the circuit.

-- Gabor

View solution in original post

0 Kudos
4 Replies
Highlighted
Professor
Professor
6,235 Views
Registered: ‎08-14-2007

Re: Update: Solved (fairly certain)....XC9500/XL driving a XR-2203 (similar to ULN2003)

Jump to solution

It's not clear to me whether these are 9500 or 9500XL (3.3V) parts you're talking about.  In either case it sounds like you have an issue with electrical over-stress.  If your XR-2203 (looks like the old ULN2003A) is driving inductive loads like

relays, you need to be very careful that the inductive kick-back doesn't get into the supply for the CPLD.  This could happen if you use a common supply for the relay drive and the CPLD.  Note that those diodes on the output of the XR-2203 dump current into the output voltage supply from the inductive loads.  If there is not sufficient capacitance to filter this reverse current you can have high voltage events on that supply.

-- Gabor
0 Kudos
Anonymous
Not applicable
6,230 Views

Re: Update: Solved (fairly certain)....XC9500/XL driving a XR-2203 (similar to ULN2003)

Jump to solution

Gabor,

 

 Thanks for the response. I thought I was clear. Apologies. It is a XC9536 (I do have some XC9536XLs, but are not being used in this specific circuit).

 

  I understand the voltage spike issue (CPLD and relays are on the same 5V supply). However, I didn't think that would contribute to heat issues on the CPLD.  However, if the inductive spike affected the CPLD (as in messed up the output circuitry), then that would make sense.

 

  Per your suggestion, I'll look at the capacitance across the supply on that board. I'll also look at the +5V line w/a scope and see what's actually going on. I wonder if I can insert some inductance in series with the relay +5 leg to see if I can reduce any spiking getting back to the CPLD?

 

  Thanks again for the input. Any other suggestions are welcome.

 

  Cheers.....

0 Kudos
Highlighted
Professor
Professor
7,409 Views
Registered: ‎08-14-2007

Re: Update: Solved (fairly certain)....XC9500/XL driving a XR-2203 (similar to ULN2003)

Jump to solution

Additional inductance won't help.  Other than capacitance, your best bet is to add another clamp diode directly across the relay coil rather than relying only on the internal diodes of the XR-2203.  That will keep any current spikes local to the relay rather than injecting them into the rest of the circuit.

-- Gabor

View solution in original post

0 Kudos
Highlighted
Anonymous
Not applicable
6,218 Views

Re: Update: Solved (fairly certain)....XC9500/XL driving a XR-2203 (similar to ULN2003)

Jump to solution

Again, I appreciate the help.

 

The relays are small (draw just a bit more than 25 mA). However, there are 4 in this design and I expect could gen a serious spike if all are activated/deactivated at the same time :(

 

Back to the design board......

 

  Thx again....

0 Kudos