Two Unrelated AVAB DDII Dimmer Questions

[kjones9999]

Hi folks.

I manage a couple of educational theatres that have AVAB DDII racks- 17 years ago we upgraded these from AVAB240 to DMX 512.

At one of these theatres, the 595-SN75157P chip goes out in 1 of 3 rack cards at least every 2-3 months. Any thoughts on the cause, or a protection against? They ran fine for 10 years or so.

I know that pins 7,8,9 on the digital db25 connector are dmx out, does anyone know what pins are DMX in?

 

[JD]

The primary reason those chips pop is when the data line is spiked with a transient. This could happen if the common/ground in the signal cable (pin #1 on a 5 pin DMX connector) was open or not making good contact in a connector somewhere. It could also happen if the DMX line was now part of a ground loop, or the DMX line is now run next to a power feed to something like a HVAC unit.
The chip's job is to take the balanced RS-422 signal (we call DMX) and translate it to a simple 1/0 data line to feed to the microprocessor. Since it is on the front end of the circuit, its the chip that gets to sacrifice itself if any external electrical abuse happens to the signal line from the board.

As for the pin-out, I would be guessing. What DMX requires is three pins. One for +data, one for -data, and the third for a common electronic ground. (Not to be confused with a frame or power ground. On standard 5 pin connectors, common is pin 1, and the data is on pins 2 and 3. (4 and 5 are not used.)

 

[kjones9999]

The primary reason those chips pop is when the data line is spiked with a transient. This could happen if the common/ground in the signal cable (pin #1 on a 5 pin DMX connector) was open or not making good contact in a connector somewhere. It could also happen if the DMX line was now part of a ground loop, or the DMX line is now run next to a power feed to something like a HVAC unit.
The chip's job is to take the balanced RS-422 signal (we call DMX) and translate it to a simple 1/0 data line to feed to the microprocessor. Since it is on the front end of the circuit, its the chip that gets to sacrifice itself if any external electrical abuse happens to the signal line from the board.

Thanks!

I have a feeling its an issue with the grounding in the building--- any "foolproof" fixes on the line itself? Optoisolator?

 

[JD]

Opto will do the trick provided you know where the fault is. If you have a board in the control booth, a run to the stage where it loops into the dimmer racks and goes no further, and the racks are tied together, then putting the opto between the board and the racks at the stage end should do the trick. If there is more downstream equipment, then the variables increase. If the opto gets nailed in a few months, then you know you have some real problems with the cable run from the booth. Some things to keep in mind:
1) Opto's isolate the input and output, but rarely isolate between outputs. So, if it has 1 in and 4 out, there is no isolation between 1, 2, 3, and 4 on the output side.
2) If you have additional runs after the rack, and they are done with 3 pin cable, make sure it is not microphone cables. Often audio cables tie pin 1 and the shell together which can cause a lot of ground problems.
3) The last device in the chain should have a "terminator" on it. Basically a 120 ohm resistor between +data and -data. Not only dose this help with data integrity, is also helps minimize voltage transients. Some racks and adapters have a dip switch to engage an internal terminator. Only the last device in the line should have the terminator.
4) Although DMX has a flow (Board to Devices), individual DMX devices on the receiving end do not have a dedicated In and Out, the jacks are simply paralleled as DMX is a "backbone" type topology.