Request for Advice - Rigol DP1308A Power Supply

[lyndonasia]

Good day all...

I was using the Red (6.0V, 5.0A) channel of my Rigol DP1308A power supply today and it decided to stop working.    I was testing some leads with a 4.0 A current at 1.0 volt.  After a couple of minutes the voltage on the channel dropped quickly to near zero.  The attached photo shows the Red and Yellow channel active.  Both channels are set to 5.0 volts and 1.0 amp.  The Red channel should indicate 5.000 volts but it shows 6 mV instead.  With a load connected the channel voltage will drop down to 0.000 V.  I ran a self-test and the two errors were:

7. +6V Out: Error
10. +6V OP: No Run

All of the other tests were passed.  When the channel failed, there was no sound, no visible smoke and I could not detect any burnt smell whatsoever.

My unit is no longer under warranty so I thought I might try to repair it myself.  However, I don't know a lot about this power supply so I am not sure if the problem would be something simple or not.  (There are no fuses on the back of the unit except for the mains.)  I have enough skill to do basic repairs and I have some gear for working with surface mount components.  I have a "service" manual but it really only shows the basics of unit disassembly.

I could take it back to the shop for servicing but I would like to have the satisfaction of having fixed it myself.  I would be grateful for any advice.

Many thanks, Lyndon

[Andy Watson]

Have you seen this thread:
https://www.eevblog.com/forum/blog/eevblog-861-rigol-dp832-psu-fail-repair/

Different version of the power supply, but similar fault under similar circumstances.

[lyndonasia]

Thanks very much Andy, I will definitely be checking that one out!  I am a bit behind on Dave's videos...  I am only up to video #461.

[lyndonasia]

Update on the Rigol DP1308A PSU Repair

Well, I just got back from a very nice vacation in Queenstown, New Zealand where I was doing some mountain biking on an e-bike!

This morning I opened up the DP1308A and had a look around.  The top board is for the two 25V/1.0A channels.  The bottom board is for the 6.0V/5.0A channel.  The little PCB fuse was still intact but upon close inspection of the power MOSFET (International Rectifier IRFP054N, TO-247) I noticed that there was a large gap of about 0.5 to 1.0 mm between the transistor and the heat sink.  The heat sink screw was also very loose and could be turned by finger.  I think this must have been a manufacturing defect as my unit has seen no vibration.  It has never been on a cart for example.  When I first got the unit I unpacked it, put it on my workbench and it has sat there ever since.

In circuit, the transistor showed a very low resistance on a couple of the leads so I decided to take it out.  My DCA55 confirmed a short on two of the pins.

I found it interesting there was no insulator strip between the transistor and the heat sink.  I checked the heat sink and it is isolated from GND but there is a bridge rectifier mounted on the other side (also with no silicone pad or mica strip).  I was thinking it might be a good idea to use a insulating strip but I couldn't find one for a TO-247 on either RS or Element 14.

When the replacement MOSFET arrives, I will post some reference values for the new versus damaged parts.  While the unit is apart I will also check the ESR on the big capacitors.  If there is anything else worth checking, please let me know.

[lyndonasia]

Replacement Fuse - What Speed?

I have replaced the pass transistor and now I want to replace the fuse.  The fuse that has blown is F2 which is a 10A axial-lead PCB type.  RS Online (HK) has a few good choices available in two speed ratings: FF (very fast acting) and F (fast acting).  The fuse is located on the low voltage AC line just before the bridge rectifier.  The channel that it protects supplies 6.0VDC at 5.0A.

I am thinking the Type F, fast acting general purpose fuse would be the best choice.  If this is not right, please let me know.

Cheers, Lyndon

Fuse Speed Ratings:
https://www.swe-check.com.au/pages/learn_fuse_speed.php

[lyndonasia]

Well, the repair is done now, so here is the final story...

It turns out there was a manufacturing defect in my unit.  The main pass transistor (a MOSFET) for the "Red" channel was not tightened against the heat sink properly so it got too hot.  It then failed with a short across its two main terminals causing a dead short to GND.  As I found out, these types of supplies often have a fuse somewhere before the bridge rectifier as a safety precaution against a transistor failure.  So typically, when a MOSFET transistor fails, you end up replacing both a transistor and a fuse.

I missed the fuse the first time around confusing it for another one next to it which tested okay.  Then for the repair, I tried a fast blow fuse but it didn't work.  It kept blowing due to the inrush current which is present when the unit is first switched on.

I then used a slow blow PCB style of fuse.  I ended up deciding on a Littelfuse TR5 Series, No. 374.  At 2.0 times the rated current value this fuse blows in about 45 seconds.  At 3.0 times, it will blow in only 2.5 seconds.  Therefore, it seemed like a good choice.

Rigol had used an axial fuse for F2.  I looked around online for one but I didn't have much success.  I could only find the more common radial style of PCB fuse.  When I tried to test fit the new fuse I found out that J1 was placed too close to F2 and the PCB fuse wouldn't fit.  (See the attached photo.)  To make more room, I moved the bleeder resistor (R16) over a bit.  I then had to desolder and remove J1 so that I could file a rounded notch on the edge of the connector.  This allowed a standard PCB fuse to fit in the space reasonably well.  I soldered the new fuse in, fired her up and lo and behold, it worked.  Yes!

The whole thing was a bit frustrating as I "fixed" the unit four times altogether.  Once for the transistor, twice with fast blow fuses, and then finally by changing to the slow blow type of fuse.  My perseverance paid off though and I got my power supplying working again.

It was a good learning experience overall, and I feel good about being able to repair a piece of commercial test equipment without a schematic, parts list or service manual! 

[tpowell1830]

You stepped through the process and persevered! Good job. 

It is a joyful feeling knowing that you faced the lion and won.

[lyndonasia]

Thanks!

[capt bullshot]

Good job and nice description of what you've done and learned!
And yes, a slow acting fuse is a good choice, the only purpuse of the fuse there is to keep the main transformer from burning in case of a catastrophic failure like yours.  In this case there's plenty of time for the fuse to blow. If I'd seen this thread before, I'd have given you this advice, sorry for beeing late.