Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion Meridian - Systems DSP33 & comms weirdness

Dear Hikers,
This is my current situation (building an Atmos system maybe 5.1.2 or 5.1.4 layout):
Processor is Storm ISP Elite mk1 with AES output option.
Fronts are DSP5200 & 5200HC legacy comms.
Rear surrounds (DSP33) being configured and installed, but having lots of problems with comms.
Volume control is is in digital domain on the storm, fixing the volume of the DSP speakers to 87 with vmsr->ip2sl->centre speaker (master & controller).
Current Comms topology:
centre ->511 "slave hub" clone
511 clone then goes
1) to the front left/right;
2) the left surround & left height speaker
3) sub (when it gets fixed)
4) right surround & heights speakers
all cables are legacy s5.
All speakers configured as type 6.
All speakers are configured as slaves except centre
The DSP33s have been configured using RS232 interface to be type 6, slave, not controller.

Here's what happens.
Centre, Left, Right all behave themselves until we plug in a DSP33. then the comms hang until we unplug it. Even if the dsp33 is not switched on, we have to unplug it and power cycle to get comms back.
With my G68ADV in a 5.1 layout using a daisy-chaining comms topology I didn't have any issues with the same 2 DSP33 I'm trying to add as rears now.


Any ideas? Looks like at least worth trying daisy chaining but why?
Could it be that the centre speaker can't drive the 511 hub by itself?
All connectors have been opened up and checked for shorts - pins 1,2,5 checked for continuity all about 1.5-3 Ohms,
Tried 2 different DSP33s , same results.

If the Storm is controlling the volume digitally anyway, perhaps it doesn't really matter, but it's bothering me that there seems to be some sort of comms issue that I can't figure out.
Other possibility is that both DSP33s have faulty comms?
Any ideas, suggestions etc welcome.

Graham

I assume that with everything in standby, You have pressed clear on the MSR to set comms controller?

Are you sure you have your comms in and outs correct?

Comms failing upon adding a DSP33 suggests a shorted buss or a comms loop.

Does a DSP33 work in place of the front left or right speaker?

Funny it seems that the DSP speaker that causes people the most headache is the DSP33

Cabling from to/from inputs & outputs all checked (several times).

From standby, Clear on MSR yields "Con" on the centre, no reaction on L&R.

IIRC on my older DSP5000 the non-controllers would show "n.con".

I'm tempted to have a look at what's going on the outputs of the suspect D33.
Suspect a short to v+ or earth locking up comms.

If I can borrow a scope I 'll have a look at the waveforms as well.
DSP33s are very inscrutable, you can't tell what's going on in there.
I have an old DSP5000C I could try in its place.

Graham

Hi Graham,

I'd try going back to basics and link from the Master speaker, and connect each pair of slave speakers one by one, and see where it breaks...

It sounds like there's either a cable fault somewhere or a DSP that has a problem with its comms.

If the 511 patch is a clone, double-check (for your own sanity) that the pins are linked correctly (i.e. 1-1, 2-2, etc.) and there are no shorts!
Double-check each comms cable - If you have a spare, prove that first and use it to replace link by link...
Are the comms cables Meridian or home-brew (if all pins are wired, that could cause an issue as the DSP33s can power 512 remote eyes, so there are volts on other pins)

Lastly, try a reset plug on the DSP33s - they can lock up and cause no end of issues that even a power-off doesn't sort!

Good luck,

Steve

Hi,

511 patch boxes are not fully wired.

Regards I

Yes, the "clone" 511 I have is all pin 1s connected; all pin 2s connected. I imagine pin 5 could also be connected together, but all S-leads will have that anyway.
The problem has now been resolved.

Turns out it was a S5 cable that had a short between pin 5 and pin 2 at the DSP33 end. This only happened when the cable was plugged in - if you tested the cable with a multimeter when it wasn't plugged in, it was ok.
I just cut the plug off, and soldered a new one using the standard S5 pinouts in the wiki.

All is behaving itself now.