Behringer DDX3216 - Trying to repair, please help!

[fznuk]

Hi,

Im new to this forum, so hi to you all!

Im a producer/composer/engineer with a home studio setup using that wonderful Behringer DDX3216 digital mixer which does have its issues...but love it!

A mate of mine also owns this mixer and has had a couple of major issues with it, mine, i havent had too many, but it has had a bit of a benny every now and then.

I have them both here now, and since i have electronics experience from past jobs, it has been sketchy but still managed to repair most things with a few pointers from other electrical engineers. I hope to have the same help and luck here if possible?

I have my friends DDX here and open to repair first, his often has the usual 'all lights on' problem but this has now been replaced mainly with just the main output master LED's staying on, no sound and unresponsive after being switched on for any given length of time.

Doing the usual read-up on the net i have come to the conclusion that the capacitors on the PSU may need changing. I have purchased and replaced all the 1000uF & 2200uF caps on just the output filter stage but on power-up i got the same problem after about 15mins.

I switched it off/on and still got the same.

I opened the PSU again and decided to reapply solder to various points on the bottom of the board, unplugged ribbon cables, plugged them back in, then switched on but this time with the case open. It lasted a lot longer....but when i finally shut the case and switched the unit back on and off i got the dreaded 'all lights on' again

Switching off and on didnt cure it, so i removed the PSU again and now checking the cables and ribbons again on the main boards, including each power connector.

Now, would it be worth it to replace all the other caps on the PSU?

Can i measure the voltages without load on this PSU without problems? and is this worth it?

Since the problems occur only after time, or random 'all lights on' occurs occasionally, would this suggest a filtered voltage problem or heat/joint problem?

Can the processor or any IC's work for one minute then not the next? Seems unlikely? but could be wrong?

Are there any other troubleshooting tests i could do to narrow down the problem?

I have two desks to fix. as i say, they work perfectly when actually working but have random glitches

Any help would be greatly appreciated while i have one desk in bits at the moment

Simon.

[km6xz]

Replacing things randomly in this unit will be frustrating and expensive since it is a complex device. The power supply needs to be checked to find if a rail is pulled down, a common cause of this symptom
There are supplies for +48, +/- 17, +/-8, +/-12, +3 and +5V. If you do not have all of these and clean, you need to find out why.
What test gear do you have available? Working with a switching power supply requires a little caution working with the secondary and a lot of caution in how you make your measurements on the AC input side.
This is a difficult unit to repair if your don't have a decent collection of test instruments.

[Enzo]

Amen, don't throw parts at it, find out what is wrong.

All peak lights on in a mixer is usually a sign that one of the analog voltages is missing, in this case one of those 17v rails.

In my own experience with these, I often find a cable from the top panel to the bottom panel coming loose. Can't tell you which one, but on the top it plugs in about the center of the panel, probably 15 pins or thereabouts.

Factory bulletin suggests C48 in SMPS dries out - 47uf/25v 105degree. I have indeed found that to be the case in some.

Can the processor or any IC work now and not later? Of course. Any part can be intermittant. More likely is the solder to it. There have been complaints that some of the LSIs on the main board were not adequately soldered. Pain in the ass, but I have seen that.

But always consider the broad picture. It is the SYSTEM that locks up, not necessarily one IC in it. Your processor may be getting a halt signal from elsewhere, and shuts itself off. A flaky power supply can certainly stop a CPU in its tracks. And all modern systems have a power-on-reset circuit, and that restarts the computer whenever powr is interrupted. However if power is glitched and returned too fast for the system to realize it went down, there will be no reset, and the system program can freeze.

[jrfrond]

This unit was also prone to bad crystal oscillators and .1 decoupling caps.

[Enzo]

And as I recall a serious pain in the ass to take apart and extract the SMPS.

[fznuk]

Thank you so much guys! and from a thread i read on here regarding this mixer, it was you guys i was hoping would reply

You are right in not throwing random components at it which is why i only replaced output filter caps which i read from a thread on another forum, but with this not making any difference, together with pulling and replacing connectors, i decided to ask on here since your other replies on posts sounded like you guys knew this unit well.

Im pretty competent with understanding electronics and with a bit of a brush up on circuits like switch mode power supplies im sure i can get my head around it but my main job is PC diagnostics so i don't come across too many component level repairs these days.

I have a purpose built electronics workshop but my test gear is limited to an old Scope and multimeter.....then usual tool...and soldering iron. But the desk is pretty big so decided to open and fix on the kitchen table!



I originally noted down all the caps on the PSU - i also noticed that its built up from three PSU's with various voltages. I was going to get all the caps but decided to try the output ones first, i will see if i have a spare 47uF to try!

You also mention the crystals - are these Q1 & Q2 on the processor board?

Thanks again for your help!

I will keep you all up to date on if i fix it or not

[fznuk]

And as I recall a serious pain in the ass to take apart and extract the SMPS.

I got to admit it wasn't too bad, was slow and apprehensive at first with all those ribbon cables, but after propping up the lid and removing the top lid connecting cables/PSU leads it was easy. Now a lot faster taking it apart and putting it back together! lol

I have another one to do yet too! lol

I think in future im going to dig out the scope and test properly, but my scope is a hand-me-down, old and maybe need re calibration?! Wouldnt mind one of these PC scopes now!

Well, gonna try changing this 47uf cap and checking the analogue 17V rails as this cap is on the same PSU as the 17V supply..... what about the 1uF cap next to it?

And, as i said in the last post, can i test this PSU without loads? Just like i can test a PC SMPS without load.

Cheers again guys!

[Enzo]

yes it should be fine without a load.

[fznuk]

yes it should be fine without a load.

All voltage rails seem to be ok, before and after 47uF change.

I have +48V (47.7), +&- 17V (16.8) and all the others. I dont have a decent scope as yet to see if there are any ripples. I have ordered one of those USB ones and decent 100KHz probe sets.

Going to roughly put the desk together again and see how long it lasts before going haywire.

I suppose i really need a scope to test the crystals too?!

I also noticed a lot of noise from all parts of the circuit... primary and secondary? I suppose the transformers are badly wound? or the coils? - is a white noise sound that's quite loud?

Maybe i should check the primary input filter caps (Large ones) and double check diodes, startup resistors and transistors?

Cheers

[km6xz]

Does the old scope work at all? If so, even if not calibrated it will be better for general pro audio use than a USB type which usually has limited bandwidth and low maximum input signal amplitude. An old Tektronix 465 100Mhz scope will cost about $150 and last 10 times longer and be more functional than a USB module.

What noise, where, how did you measure it....I am confused. You already determine the supply is not causing the mixer to be locked up so move on to more logical causes. There are lots of crystals in the mixer, 24mhz, 12.xxxmhz etc. Your meter probably can't see that but you can see if data lines are steady state "1" or "0", by reading a DC level of some where between 5volts and zero volts.
Where the supply voltages measured with the cables to the rest of the mixer installed or without a load? If no-load, try hooking up the ribbon cables and measure again to see if one rail is dropping out under load. If your measurements were with the load of the circuits, forget the supply and assume it works.

[jrfrond]

In an effort to help here, I just checked every DDX3216 repair in our database. Overwhelmingly, the solution we took to the problem described here was the replacement of crystal Q3, 32.768kHz. Over and over, the problem reads the same: runs for awhile, then locks up, no control, etc. Sure, there are also the ubiquitous cap issues and poor soldering, but this fits the bill. Hey, it's worth a shot. You need to eliminate this common issue first. We stock these crystals. Mouser has them as well, part# 559-NC38-LF.

[fznuk]

Does the old scope work at all? If so, even if not calibrated it will be better for general pro audio use than a USB type which usually has limited bandwidth and low maximum input signal amplitude. An old Tektronix 465 100Mhz scope will cost about $150 and last 10 times longer and be more functional than a USB module.

What noise, where, how did you measure it....I am confused. You already determine the supply is not causing the mixer to be locked up so move on to more logical causes. There are lots of crystals in the mixer, 24mhz, 12.xxxmhz etc. Your meter probably can't see that but you can see if data lines are steady state "1" or "0", by reading a DC level of some where between 5volts and zero volts.
Where the supply voltages measured with the cables to the rest of the mixer installed or without a load? If no-load, try hooking up the ribbon cables and measure again to see if one rail is dropping out under load. If your measurements were with the load of the circuits, forget the supply and assume it works.

hi, and thanks again to you all for help with this mixer! much appreciated!

Yes, the scope does work:



Its a Gould 12Mhz one.... given to me a few years ago.

The USB one is just to have a play with really, and cheap so no love lost if its no good....well i will see when i get it!

I have also ordered a Cap test meter which will help...i think! lol

After brushing up on a bit more with regards SMPS, i feel i have learnt a lot since embarking on the mixer a couple of days ago. being absorbed into electronic reading material has never really been my strong point...more of a practical person....and occasionally blowing things up to learn type! lol

The noise was always there, a kind of hissing, but seemed to get worse across the board?! I think coming from the transformers/inductors?! - But, on powering up the mixer with it (top wedged open and PSU without a lid) the desk fired up....and got a different light show..... but then a flash of sparks from the PSU area..... As i rushed to power off I noticed that one black PSU lead had dropped from within one of the PSU plugs going to the top back right of the console. I removed the white plug and reinserted the lead but when i plugged it back into the PCB it fell out again as the plug was inserted. This time i re-inserted the lead/pin and made sure it was secure........ plugged the desk back up and it worked a treat. I attached powered speakers and a sound source and left it on, i switched it off and on a few times (something this desk hated in succession) and removed the sound source 1/4" jacks and reinserted them, again, something this desk loved to 'crash' on, but it worked well for at least 2-3 hours.

After this i removed the sound source and speakers, just about to remove the PSU to box it up and thought i'd just turn it back on again....bang.....i got the all the lights problem?!! I switched it off and straight back on.....the problem went away and the desk was back (something the desk hated when it got the lights of death as you would have to wait 20-30 mins to switch it back on).

So... am i getting somewhere or going round in circles?! - i suppose this is where the scope comes in!!?

In my line of work i dont usually need a scope and havent used one in a long time...but im sure its like riding a bike and my training will come back! and prob never put the scope down again!

I do get electronically lazy at times but determind to fix this desk....the right way!

And will probably get my head back into electronics books again to brush up on my old electrical repair knowledge!

...............shamei cant get as excited in getting back into music composition! lol

Crystals - I will look into this as long as i know the PSU is steady as i tested it without load. So next i will check with load and see if any rails drop.

Cheers guys!

[fznuk]

In an effort to help here, I just checked every DDX3216 repair in our database. Overwhelmingly, the solution we took to the problem described here was the replacement of crystal Q3, 32.768kHz. Over and over, the problem reads the same: runs for awhile, then locks up, no control, etc. Sure, there are also the ubiquitous cap issues and poor soldering, but this fits the bill. Hey, it's worth a shot. You need to eliminate this common issue first. We stock these crystals. Mouser has them as well, part# 559-NC38-LF.

Thanks for that, but im in the UK so maybe i should have a look here and see if i can get the crystals for a decent price.

One other thing, if the primary large caps were on the edge of failing, would this mixer work but cause lockups like this?

With the scope i suppose i can check for ripples?

Cheers!

[Jazz P Bass]

You can check Vdc for "ripple" wih a decent voltmeter.
Set the meter to read Vac.

[jrfrond]

Thanks for that, but im in the UK so maybe i should have a look here and see if i can get the crystals for a decent price.

One other thing, if the primary large caps were on the edge of failing, would this mixer work but cause lockups like this?

With the scope i suppose i can check for ripples?

Cheers!

The crystal is made by Fox and their part# is NC38. You should be able to source it in the UK. It's a small watch crystal oscillator. Seriously, you NEED to eliminate this. I COULD be wrong, but according to our repair documentation, there's a good chance I may be right about this.