Ensoniq DP/4+ REPAIR (cap leakage, corrosion)

[Script]

And wrong I was again.

U1 swap went easy-peasy, including cleaning pads. All pin measurements 'reasonable' afterwards.

U101 however suddenly had around -5V on IN pins (CH3). Thought it was cos of diode(s). Anyway, I took that op amp out too -- better anyway for symmetry. Turns out the diodes around it look fine on DMM (but will order batch anyway).

Then when trying to clean U101 pads, some came off just like that at the slightest touch of solder wick accompanied by evil burnt "acidy" smell. Yappari ! That IC sits close to the worst affected part of the input section. (Image top left: some green paintwork covering fixed trace.)


Half-cleaned electrolyte swamp underneath IC. -- Bad !



And now I also get to fix wrecked SMD IC pads. -- Good !

 

[ccaudle]

now I also get to fix wrecked SMD IC pads.

One trick our lab techs used to use when a pad got ripped is solder the legs to the remaining good pads, then solder a 30 ga. wire wrap wire (the kind with kynar insulation, very heat resistant so it doesn't shrink back when soldering) to the top of the leg with no pad underneath, and route that wire down flat against the PCB over to a via or component lead to solder. You are effectively replacing the trace on the PCB with a small wire. I had to go look this up, but Kynar is the branded name for generic PVDF insulation. It is definitely superior to PVC or other rubbery type insulation when you are trying to solder on short lengths of wire between components.

 

[Script]

Thank you. I'll look into those wires. Sounds like something very useful to have around. Insulation when running inbetween components is paramount. I was thinking jump-wiring too. Probably the safest bet in terms of time and budget (if applicable). -- Oh, I just saw that I do have such insulated wire here. Didn't use it though. Instead I soldered wires onto traces, which of course takes more time.

In image above, four damaged pads:
2 and 6 are completely gone
3 is loose but still somewhat hanging on the two traces
4 is loose and a tiny part of pad came off (probably OK)
(7 is fine,-- dark part on top is just shade)

Have been watching many SMD repair videos on YT lately. I generally like what people in India do and upload -- they can be very resourceful. What I've been doing so far is shown in several videos uploaded by a guy who is fixing IC pads in following video from around minutes 21:30.

 

[Script]

Tiny steps. Not much time at the moment.

Pin 6 of U101 hadn't connected before, so I had quickly jump-wired it with copper wire from IC pin to two components. That was before even taking out all electrolytic caps. It looked like this.



Turns out I like nail polish after all. It probably doesn't make for much added stability but it's easy to remove by simply scratching it off the board..

This time, for this IC, I soldered wires right onto traces again and so bridged the missing pads. Took pad 3 out, cos it was hanging loose already anyway. Pin 2 has a "pad" created by folding wire into a closed hook soldered together. Heated pad 4 and it didn't come off, so I simply placed a blob. Pad 7 is fine, just a shade.



Also found another 'eye' that looked bad (top left) . It was still filled with electrolyte, and the veer was gone on board underside with gunk oozing out of the 'eye' when heating it. Let's call it an electrolyte pimple.

Time to install the IC and fix the overlooked 'eye' on top left with wire through the hole and onto the trace.



Had green-coated the wires underneath before installing IC. Wires between other componets sit tight cos soldered onto traces. Coating them without drowning adjacent components is a job for the dentist again ; )



This approach takes longer than just wiring from pins to components, but should the IC burn out again, I thought, it should be easier this way to desolder and replace. Alternative would be green insulated wires onto pins and traces.

##
Voltages on all pins of U101 (CH 3&4), as well as on U1 (CH 1&2) look much better now. Still some 0.08V on input and output pins of U101 (but 0V now on IN and OUT pins of U1 where it had been 0.5V before with only U1 replaced). Supply voltages also shifted from +15.67 to +15.55 and -14.88 to -14.98.

According to service manual, another way of checking whether there's a faults on this board is to check for ideal voltages ((+/-10%)) on jumper 3 -- juice coming from power supply but before regulators on this audio board. I definitely have too low before regulators and with +15.55V probably still a tad to high after positive regulator.

-- The search continues. --

 

[flow808]

Hi,

I have a DP4+ that works well, but I'd like to perform a preventive maintenance as I understood cap leaking can damage it badly.

I saw a kit on eBay ENSONIQ DP4+ COMPLETE REFURBISH DIY KIT FOR THE TECH SAVY!!WEEKEND ONLY SPECIAL! | eBay and I'm wondering if it is worth it, or if the price difference would be big enough to open it myself and to list the different capacitors and voltage regulators and order them.

Did anyone here have any experience with this listing? Is there by chance any BOM already available ?

 

[Script]

Welcome to the forum.

//
Don't have a complete BOM here. I get what I need as I go along.

This unit operates producing extremely high regulator heat (+/-29VDC regulated down to +/-15V and +/-5V plus -13VDC down to -5V). Already BBQ with lid still off.

Therefore I'd advise swapping out all electrolytic caps in the unit on all four boards and both SMD and thru-hole (on PSU).

SMD caps are really not expensive. I chose higher voltage-rated, cos same was not available. Need correct footprints though (three different uf values = three different sizes -- height doesn't matter). Didn't care much about maker cos any SMD cap today is likely to be better than what they used back in 1994/95.

Bigger heatsinks crossed my mind (maybe) or just a few more ventilation holes. But this later, if ever at all.

 

[Script]

Very slow progress.-- almost no time to spare.

Red overload LED of Channel 3 on frontpanel was constantly on. Suspected the reinserted trimmer and/or fixed traces around there. It looked like this.



The trimmer had measured fine before reinsertion. But I had broken one of my own rules, never to reuse components unless unobtainium. Well, this one a cheapo single-turn trimmer -- just why did I do this ?!

When putting this original trimmer back into place, I had to solder leads on topside of board underneath the trimmer. I know for sure I did not touch any of its plastic part with the iron, but the heat from the iron must have affected the trimmer's plastic axis. Result : wiper didn't connect any more. Took trimmer out again and inserted a multi-turn with leads bent to make it fit. Like multi-turns more anyway. Connectivity now good and same resistance dialed in as before (fine-tuning for later). Looks nice ? Who cares.




Still, overload LED on Ch 3 wouldn't turn off. So I measured all pins on all ICs, and using the schematic, traced all voltage readings through the circuit to identify the faulty area. Could have been one of two ICs. Feedback on one of those IC amplified the wrong voltage correctly.,so I replaced the other one:, U19, and bingo : voltages on channels 3 and 4 OK.

But now, corresponding opamp for Ch 1 and 2 was off. Weird. It either suddenly also failed or this is because it sits further away from the regulator. So I also swapped U16 out. Had time to quickly measure all non-rail pins on those ICs and they read fine, except for the one with the trimmer, which is for DC voltage offset : )

---
So far four blown dual opamps. They all sit close to electrolytics that had leaked and heavily corroded traces.

---
Still waiting for diodes to arrive., all 1N914 type.

 

[Script]

Little to no progress. Drowning in work right now. But I have a favor to ask., cos I don't have access to a working unit to compare to -- which would make troubleshooting much easier.

Anyone has a unit open right now and would mind doing some quick meaasurements for me ?

I'm looking at ICs U17 and U55 (CX20018 AD converters) and am not sure what the readings should be here on the INs and OUTs. Specifically:

U17
pin 1:
pin 2:
pin 7:
pin 8:

U55
pin 1:
pin 2:
pin 7:
pin 8:

That would help a lot. I am not in a hurry with this one.
Thank you.

 

[Whoops]

Anyone has a unit open right now and would mind doing some quick meaasurements for me ?

I have the same unit, I’ve been wanting to open it up since I’ve seen the damage inflicted in your unit. I need to check if there’s any leakage going on in my unit.
So I will open it and take the measurements you asked also

 

[maxbobo]

Does anyone who has been working on these have a BOM or list of capacitors for the DP/4+?

I have a unit that has very noisy outputs. It came from a seller who said he had done some work to it, replacing the CX20018s and caps around its inputs, as well as a few opamps, and some work on the power supply. I want to finish what he started - since the caps were in questionable condition they are probably all ready to go soon.

The problem is that I don't know exactly what was worked on and what wasn't. There isn't any visible corrosion.. yet. So I thought I may as well just do a full recap.

I am not used to dealing with or ordering smd packages so if someone had already gathered this info it would be much appreciated.

 

[Whoops]

Does anyone who has been working on these have a BOM or list of capacitors for the DP/4+?

There's some schematics available at this page, also if you search google you will find a service manual plus other info:

https://convexoptimization.com/EnsoniqDP4/
Probably Script will have all the docs that are available since he is doing a pretty deep and difficult repair job

 

[Script]

My unit is at end of the queue at the moment, cos I would like to confirm some voltages before throwing in more serious money for ICs.

Your unit is open ? Does it work, other than for the noise ? Would you mind measuring it for me (see couple of posts back)? Could. look for what I ordered. and used in repair so far. Also, noise might be due to some 0.1 bypass caps, I had read, and therefore exchanged those caps (just in case and for practise ).

 

[maxbobo]

It is working other than the noise, with one exception. The LEDs on all outputs will peak Red even when incoming audio is at a useable (green) level. User sees a green LED the red LED at the same time while watching the metering.

The unit isn't open. I re-racked it to work on layouts a bit. I want to do some midi CC modulation from my daw so working on learning that until I get it fixed.

I'll read back so that when I have it out later this week I can offer some helpful measurements.

 

[j.colt]

FWIW my DP4+ had the symptom you described before finding out that the SMD caps had leaked and corroded the traces and vias beyond repair--at least I was not brave enough to attempt the epic repair discussed in this thread! I suggest you stop using your unit immediately and start replacing caps before they eat into the pcb too much.

 

[Script]

All documentation I have I uploaded to Technical Documents section, including Service Manual. As described in it, could run on-board diagnostics program to get some clue for deducing what's wrong.

 

[maxbobo]

I removed the caps on the analog board to check for corrosion.

There was definitely evidence that it was beginning, as attached, but every trace I checked out point to point had continuity.

So I went ahead and shotgun approach - recapped the whole analog board. This has solved the problems of peak LEDs lit every time any audio is applied to the inputs, and the outputs of channels 2-4 are quiet.

But now... Channel 1 has a permanent green LED and some hissy white noise when the input is cranked. The unit is kind of usably working though otherwise.

I've still got the digi board and psu to do, but I would guess this is my fault. I've done barely any work with SMD components and maybe damaged a trace or shorted to a via in the process...

 

[JohnRoberts]

look for a solder bridge...

JR

 

[Whoops]

After seeing this thread I’m really scared to open my Ensoniq DP4+…

last time I used it, 1 year ago, it worked perfectly but I’m sure it will eventually develop the leakage and corrosion showed that has been been shown here
I guess I really should do something before it’s too late.
I really Hate working with SMD

 

[maxbobo]

Everything looked good with a jewelers loupe - I couldn't find any bridges so far.

Sure enough, I'm getting Bad DC Offset now, and the same symptoms on Ch. 1 but with constant noise. Looking around th

e 4565 on the way into the converter... maybe input op-amps are shorted, though I would expect to see this symptom on Ch. 2 if it were the ic itself as well since they are shared.​

 

[Script]

Inputs are four identical channels. On PCB, only the component placement for channel 1 input differs slightly . Otherwise identical. Makes it easy to measure channels for comparison and fault finding.

Measuring and comparing all dual opamp IC pins should reveal faulty area.

Or converter needs recalibration: look at the trimmers on schematic. That's where I got stuck, cos I have a feeling, the converters on my unit are over the styx. (But on mine needs confirmation through comparing with working unit.)