Strange Hardware Problem.

[tkvenki]

HI All,

I have a strange hardware problem while connecting my interface card to my backplane. The interface cards have CPLD's in them. Once my Master interface card sends a address to the common bus, the interface card which has a matching address will respond by a acknowledgment.

The acknowledge pin of all the interface cards are programmed to be in High Impedance initially. They are pulled up in the back-plane ( as shown in the circuit attached). Once the address sent by master is matched with the hard-coded address of the slave-interface card, the acknowledge pin of that particular interface card is made low by the salve card.

But if I observe the waveform using an oscillator, the amplitude of the acknowledge signal is attenuating for every interface card i insert into the product. I mean, if i insert 1 card then the ack signal amplitude is around 4.5V. If i insert the second card, the amplitude of the acknowledgment signal is 4V. Another card insertion will bring down the voltage still!!!!!
Because of this I'm not able to have more than 4 cards in my product.

Please help me out in getting through this strange issue.......!!!!!
Please find the circuit attached.

 

[Pommie]

It sounds like you're trying to implement an open collector type interface but the outputs aren't going to high impedance. Can you change R1 to a diode pointing left and get rid of C1?
I'm assuming that R1 is repeated for each output.
Mike.

 

[bananasiong]

Why is the diode needed? Just remove R1 and C1 will do?

 

[Pommie]

Why is the diode needed? Just remove R1 and C1 will do?

From the OP's description, it sounds like the outputs are not open collector but are driven active high. The diode will make it act as open collector.

Mike.