Behringer DSP-8024 Help Needed

[Tejaus]

Hello,
I have a DSP-8024 that won't boot up past the "Welcome' message. In addition to the message it will flash the various LED's back and forth. One of the large power supply caps was bulged so I replaced all three of the large ones and a few of the nearby smaller ones even though they tested OK.

I've tried three different backup batteries with no luck. Are there know issues with the PS on these? I'm thinking it maybe the EEPROM? The firmware is
ver. 1.3. Are there a 1.3a and 1.3b revisions? I've seen both mentioned on the Internet.

I know there is a protection mode with the relays. They never kick in on mine. I've also looked for cold solder joints etc.

[Enzo]

First check power supplies. You replaced bulged caps, but do each of those caps have a nice smooth DC voltage? The relay doesn;t kick in because it never finishes its wakeup sequence. My first suspect is excess ripple or low voltage on one of the computer voltages.

[km6xz]

Did the backup battery need replacing? If not, why risk introducing more unintended problems before figuring out the real problem?
Guesses in the dark and swapping out good parts will only make repair harder and less likely. Don't change anything unless you can prove that it needs it. Do as Enzo says, check the power supply output for voltage and ripple on a scope, it is the most likely cause of this problem. Leave the EPROM's alone, it is extremely rare for one to fail, it is NOT a version issue.
Verify the clocks are of the right amplitude and frequency with a scope.

[Steve Conner]

Did you try the factory reset procedure? If it didn't need a factory reset before, it probably will now that you've messed with the backup battery.

[S2udio]

Leave the EPROM's alone, it is extremely rare for one to fail, it is NOT a version issue.
Verify the clocks are of the right amplitude and frequency with a scope.

+1

And double check all the PSU voltages first,
As it seems schems are rare or non existent.....some inside pics could help .

[Tejaus]

OK, checked for AC Ripple and DCV.

Out of the PT are two sets of secondaries. One is center tapped for the 7915/7815, the other is a pair marked on the PCB "AC" and "Cm" for common I suppose. This set goes to a LM317

AC voltages out of the PT. 20.24/20.24vac for the CT, 6.06vac/0.50vac the other.

7915: G=0, Vin=-25.76vdc, Vout=-15.06vdc with 129mv AC ripple on Vin.

7815: Vin=24.00vdc, G=0, Vout=15.23vdc with 50mv of ripple on Vin

LM317: ADJ=3.84vdc, Vout=5.1vdc, Vin= 13.25vdc with 1.35Vac ripple.

The Vin on the 7915 shows continuity to ground sometimes then a cap will charge the meter. The three legs of the CT secondary at the PCB also show continuity to ground. Leaky 7915 maybe or the IN400x diodes?

Now I did manage to get it to boot all the way up and fiddle with the programs etc. This was after I had pulled the board and replaced two small 10uf/50V bypass caps just for grins. They measured OK on my cap meter.

I installed the board back with just one non grounded PCB screw, the the three screws holding down the regulators to the heat sink and the PCB to chassis ground wire.
I let it run for quite awhile then prodded the ground wire with a chop stick and it went back to the protection mode. I checked that wire for continuity. It looks OK.

Per the manual pulling the battery is how these are reset. It will still work without, you'll just lose your settings.

I'll get some pics up soon.

[Enzo]

Your 15v supplies produce 15v, so the transformer is working. The leads from the CT winding show continuity to ground because the CT is grounded, and the winding is wire after all. That is OK.


If poking a wire causes a symptom to either start or stop, then that poking esxposed some loose connection. It may not be that wire itself. When you poke the wire, you also put a little yank on th circuit board.

[Tejaus]

Thanks, that narrows it down a bit. I double checked the ground wire. sprayed a little Deoxit on it too.

I'm going to clean the ground pads on the pcb tomorrow. And the tops of the standoffs.


What about the 7915 Vin leg sometimes showing ground?
And the 1.35vac on the LM317?

[Enzo]

The 1.35AC is ripple on the INPUT to the 317.

You have a circuit full of capacitors, and when you measure resistance with your meter, the current the meter uses tries to charge the caps around it. The less charge there is in a cap, the more it looks like a short circuit. Plus polarity matters. If you try to measure resistance and there is a tiny residual charge voltage in the circuit, then your meter curent has to first discharge that voltage and then charge it up the other way.

Grab an electrolytic cap from your parts supply. Measure its resistance with your meter. It will start low, then slowly go up as the cap charges. Now swap ends with your meter leads. I bet your meter starts out showing "negative resistance." SOmething meaningless like -12k ohms. That slowly drops to zero, then it starts climbing with regular resistance readings. Another paret of this experiment is to measure resistance of thgis cap until it stops going up. Now flip your meter over to DC volts and see what is now stored in the cap.

If your three power supplies are making their nominal voltages and clean, and they do it every time you power up, then I wouldn;t worry about meter anomalies.

[Tejaus]

Well I re-flowed the solder on the PCB ground pads and made sure the PCB ground wire was firmly attached to the chassis. It worked for awhile and I left it running for about ten minutes. I could turn it on and off. I then turned it off to let it cool down and now it's not working again.

Any idea on what the fault protection circuit is on this?

From the Owners Manual:

Fail-safe relays have been incorporated into the design of the BEHRINGER ULTRA-CURVE PRO, which
automatically and silently bypass the unit in the event of power supply disconnection or failure. These relays
are also active at switch-on to isolate the ULTRA-CURVE PRO until the power rails have settled, thus preventing
the possibility of a potentially damaging switch-on thump.

[km6xz]

Maybe I missed it but before resoldering, did you check if the 3 clock oscillators are running? The intermittent nature of the problem sure sounds like a crystal oscillator not firing up each time. The phase shift chip caps on the crystals are suspect but occasionally it is the quartz crystal itself. Check with a high Z scope probe and note the relative amplitude. If one is much lower than the others it might be running on the wrong overtone.

[Enzo]

The relays connect input and output jacks, they have nothing to do with making it wake up when powered.

[Tejaus]

OK I put a probe on the the three crystals. The one on the far left doesn't show anything, the other two do. When I put the oscope probe on the far right one the LED's will halt flashing until I remove it.

There are three different frequencies. Raltron branded.
I think this is the right spec sheet:
http://www.raltron.com/products/pdfs...c_49_45_51.pdf

The one that is not showing a signal is marked:

Raltron
11.2986-20
02NK38

Middle one:
Raltron
12.288-20
02NK38

Right one by the PS:
Raltron
12.000-20
02NK05

[Tejaus]

I can find the 11.2896 Mhz crystals at Mouser but not in the HC-49/U case, just the shorter version http://cfm.citizen.co.jp/english/pro.../HC-49_U-S.pdf

Would that matter spec wise?

[olddawg]

OK I put a probe on the the three crystals. The one on the far left doesn't show anything, the other two do. When I put the oscope probe on the far right one the LED's will halt flashing until I
02NK05

Without a flow chart it is hard to tell, but that third crystal may not start running until the system wakes up. I would resolder that crystal being aware that is probably a double sided board. Once that is done, if it doesn't work, tickle it with some freeze mist. If you have a pinout on the micro look for the reset pin. Is it a - or + reset. You can usually do a manual reset by momentarily grounding the pin through a 1k resistor or to the 5v supply. You need 3 things: power, clock, and reset. If you have a scope you monitor the reset pulse.