Author Topic: EEVblog #512 - Rigol DP832 Bad Design Investigation  (Read 322059 times)

0 Members and 2 Guests are viewing this topic.

Offline Corporate666

  • Supporter
  • ****
  • Posts: 2008
  • Country: us
  • Remember, you are unique, just like everybody else
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #600 on: June 04, 2014, 04:21:50 am »
If any company won't/can't manage international issues, why the hell they choose to sell internationally? Oh, yes... maybe because profit. We're all great merchants when all goes fine. It's when you must manage troubles that you show your company reliability.

...and the terms are clearly spelled out when you make a purchase.  If someone chooses to buy internationally to save money, they take on the liability of dealing with those ToS.  That means if the shit hits the fan, you might have to pay return shipping.  It is the vendor who sets the ToS, and it is the customer who chooses to shop there or not.  If you are selling your car and I buy it and ship it to the USA and then have a problem, is it now your responsibility to ship it back to your home country to be fixed at your cost?  Of course not. 

Quote
You don't want the international customers to bother you? Stop selling internationally. Period.

Probably in your country. Not in EU for a thing like this one (bad engineering).

You have no experience running a business.  If you buy something from another country, you don't have the benefit of whatever local laws and regulations govern transactions.  Like I said above - if you want those protections, then buy locally.  If you buy from another country to save money, you take a risk.  When things go wrong, that's just the price of taking that risk.  Sometimes shit happens. 

We all have to put our big boy pants on at some point in our lives and take personal responsibility for our actions - including accepting the ToS of stores we choose to shop in. 

It's not always the most popular person who gets the job done.
 

Offline Bomber18

  • Contributor
  • Posts: 12
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #601 on: June 04, 2014, 04:52:22 am »
So now you've determined the whole logic board will fail within 2 years based on.... what exactly?  I am not aware of any failures - are you?  And... Rigol offered to fix the problem.

Well what do you think will happen when you expose those two capacitors millimeters away from a part that hot over time? Not to mention that the part itself is operating at the tipping point of it's thermal limit  I don't think calling the thing a time bomb is too far off myself.

Yes Rigol "offered" to fix the problem by posting a website for RMA requests. Unfortunately I submitted my information and heard nothing for 6 months and it doesn't sound like I was an isolated incident. Also a 6 week turn around time is pretty absurd for those that were able to get RMAs fulfilled.
 

Offline ivaylo

  • Frequent Contributor
  • **
  • Posts: 661
  • Country: us
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #602 on: June 04, 2014, 06:18:42 am »
Corporate, for whatever reason you are taking this too personally. If you are so happy with Rigol & Tequipment just say so and let's move on. I am in the states, not international or anything, I still spent a year in limbo plus probably 20 to 30 hours of my time dealing with this. Then I got my money back. Now let's hear what experiences others had...
 

Offline mcinque

  • Supporter
  • ****
  • Posts: 1129
  • Country: it
  • I know that I know nothing
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #603 on: June 04, 2014, 11:56:13 am »
So now you've determined the whole logic board will fail within 2 years based on.... what exactly?
Well, a regulator can't live for many hours at that temp. An of course the caps near him have a limited life. It's not rocket science.

Quote
I am not aware of any failures - are you?  And... Rigol offered to fix the problem.
Oh, yes. It's only that they doesn't reply to me and my local distributor just ignores me.

Probably in your country. Not in EU for a thing like this one (bad engineering).
You have no experience running a business.
And you think you know everything about the others just reading two lines.

I was referring to EU purchases. I live in EU, and I've bought in EU.
 

Offline Rasz

  • Super Contributor
  • ***
  • Posts: 2616
  • Country: 00
    • My random blog.
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #604 on: June 04, 2014, 02:28:51 pm »
Fact is, there was a part running too hot, so they fixed it.  Next.

and it only took them few months of shipping back and forth

Did they by any chance fix channel two turning on output by itself? and bad readings?
Who logs in to gdm? Not I, said the duck.
My fireplace is on fire, but in all the wrong places.
 

Offline tequipment

  • Regular Contributor
  • *
  • Posts: 207
  • Country: us
    • TEquipment
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #605 on: June 07, 2014, 04:25:40 pm »
Hello.  Just to be clear.  If anyone has had an issue with any item contact me and we will take care of it.  We have a staff and they have procedures.  After 30 days the issue is a warranty issue with the manufacture.  This is the policy BUT that dosent mean we cannot help.

Send me a message and I will have someone follow up on it.  We stand behind what we do.  I have owned this company for 12 years since I got out of college as an EE.  We have the integrity and we stand behind our customers.  After all our rep is everything. 

I am out next week but will forward anything I get.

Have a good weekend.  Off to clean the yard and go to the kids soccer games. 
Thanks
Evan Cirelli

Vice President and Co-Founder
TEquipment.NET
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #606 on: June 07, 2014, 05:03:42 pm »
If anyone owns a DP832, can they test the parallel configuration of this PSU and report back?

Are both channels stable when they are connected in parallel?  What happens if you make one channel differ by 1 V?  In theory, the channel with the lesser voltage setting becomes the output voltage of both channels, and the channel with the higher voltage setting will match that voltage setting and go into constant current mode.  This is a stress test, but any linear CVCC supply can perform that test without instability [ albeit not advised in the long term as one channel is 'overworked', ideally each channel shares the work equally.]

The Rigol application note:

http://www.batronix.com/pdf/Rigol/DP8_DP1_ActiveLoads.pdf

States no 'active loads' should be applied to any channels as it could damage the 'powered device.'  It implies its not advised to parallel even its own other powered channel.

But what I think Rigol means is you cannot connect a fixed voltage source to any channels such as a fixed voltage wall wart supply, a battery or a solar cell.  A CVCC device adapts dynamically until an equilibrium is reached, and both devices are in a steady state.

Finally, an 'active load' means something like an electronic load, which clearly can be used on the DP832 so I think the wording they chose is not ideal.
« Last Edit: June 07, 2014, 05:05:35 pm by saturation »
Best Wishes,

 Saturation
 

Offline Rigby

  • Super Contributor
  • ***
  • Posts: 1476
  • Country: us
  • Learning, very new at this. Righteous Asshole, too
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #607 on: June 07, 2014, 05:33:43 pm »
The first page of that document explains what they mean when they say "active load" - any device that can source power into the supply.

When running channels 1 & 2 in parallel, I've always had them "ganged" so that controlling one controls both simultaneously.  I've not had an issue with that.  That doesn't answer your question, I know, but it might help knowing that you can set up a master/slave channel arrangement so that you don't need to worry about setting the channels independently.
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #608 on: June 09, 2014, 02:37:27 pm »
Much thanks Rigby.

Since you got it working with the Rigol's track function [ which 'gangs' the channels] I take it the parallel setting works ok but one should test it for stability to make a claim it supports parallel mode. 

When you set it to track and parallel the outputs do both channels stay in CV mode?  That would be excellent. 

It should not switch between CV and CC mode alternatively under load as it would cause DC regulation problems.  To stabilize it one channel is set 0.25-0.5V higher than the other and permanently sets one channel as CV and another as CC.
 
Most CVCC supplies allow any other CVCC supply to be paralleled into its ports so you can take 'N' linear supplies and output 'N' current.  The difference in voltage between either channel can be ~1 V without damage so one can adjust the channels to balance them manually. 

The first page of that document explains what they mean when they say "active load" - any device that can source power into the supply.

When running channels 1 & 2 in parallel, I've always had them "ganged" so that controlling one controls both simultaneously.  I've not had an issue with that.  That doesn't answer your question, I know, but it might help knowing that you can set up a master/slave channel arrangement so that you don't need to worry about setting the channels independently.
« Last Edit: June 09, 2014, 02:41:20 pm by saturation »
Best Wishes,

 Saturation
 

Offline Rigby

  • Super Contributor
  • ***
  • Posts: 1476
  • Country: us
  • Learning, very new at this. Righteous Asshole, too
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #609 on: June 09, 2014, 05:24:30 pm »
The user's guide for that power supply specifically mentions parallel connections between channels 1 & 2, and I've done it myself.  I've not gone through the motions, I just expected the PS to behave, and it did.

I've also used the track function to supply an adjustable positive and negative voltage from channels 1 & 2.  that worked great, also.
 

Offline thmjpr

  • Regular Contributor
  • *
  • Posts: 175
  • Country: ca
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #610 on: June 12, 2014, 08:10:47 am »
If anyone owns a DP832, can they test the parallel configuration of this PSU and report back?

Are both channels stable when they are connected in parallel?  What happens if you make one channel differ by 1 V?  In theory, the channel with the lesser voltage setting becomes the output voltage of both channels, and the channel with the higher voltage setting will match that voltage setting and go into constant current mode.  This is a stress test, but any linear CVCC supply can perform that test without instability [ albeit not advised in the long term as one channel is 'overworked', ideally each channel shares the work equally.]

I don't see how what you are describing is possible unless the supply is four quadrant (or has a crowbar circuit?). What I expect to happen is the voltage output will be whatever channel has the higher voltage, as it boosts the lower channel up. But I will describe what I noticed.

As soon as I connect CH1 to CH2 there is a 40mA current that passes between channels. Presumably this is a constant current load activated when O/P voltage exceeds set voltage, to drain the caps. I can set CH1 to say 10V, and CH2 can be set from 0-10V with the same 40mA draw (and both displays read 10V as its o/p is boosted by the other channel). However, the opposite is not true, if I set CH2 to 10V and CH1 to 0V, the display reads 0.3A on CH1 and 40mA CH2. I'm not sure if this is a software bug (from reverse current flow), as the external current flowing is still 40mA. Or maybe circuit architecture.

One thing I thought though, if you do a lot of paralleling it would be nice to have the supply auto-add the two channels together on the display.
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #611 on: June 15, 2014, 09:47:16 pm »
Thanks thmjpr for the detailed reply.  The basic response of a CVCC supply in parallel is well described; here's a good reference for the details:

http://www.home.agilent.com/agilent/editorial.jspx?ckey=520808&id=520808&nid=-11143.0.00&lc=eng&cc=US

http://powersupplyblog.tm.agilent.com/2012/11/how-can-i-get-more-power-from-my-power.html

When you adjust one channel's voltage higher than the lower voltage set channel do you notice it switch from CV to CC annunciator on the LCD screen?

The response you write off for the PSU is a bit odd, I'm analyzing it now and will report back.


If anyone owns a DP832, can they test the parallel configuration of this PSU and report back?

Are both channels stable when they are connected in parallel?  What happens if you make one channel differ by 1 V?  In theory, the channel with the lesser voltage setting becomes the output voltage of both channels, and the channel with the higher voltage setting will match that voltage setting and go into constant current mode.  This is a stress test, but any linear CVCC supply can perform that test without instability [ albeit not advised in the long term as one channel is 'overworked', ideally each channel shares the work equally.]

I don't see how what you are describing is possible unless the supply is four quadrant (or has a crowbar circuit?). What I expect to happen is the voltage output will be whatever channel has the higher voltage, as it boosts the lower channel up. But I will describe what I noticed.

As soon as I connect CH1 to CH2 there is a 40mA current that passes between channels. Presumably this is a constant current load activated when O/P voltage exceeds set voltage, to drain the caps. I can set CH1 to say 10V, and CH2 can be set from 0-10V with the same 40mA draw (and both displays read 10V as its o/p is boosted by the other channel). However, the opposite is not true, if I set CH2 to 10V and CH1 to 0V, the display reads 0.3A on CH1 and 40mA CH2. I'm not sure if this is a software bug (from reverse current flow), as the external current flowing is still 40mA. Or maybe circuit architecture.

One thing I thought though, if you do a lot of paralleling it would be nice to have the supply auto-add the two channels together on the display.
« Last Edit: June 16, 2014, 12:02:07 pm by saturation »
Best Wishes,

 Saturation
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #612 on: June 16, 2014, 12:15:09 pm »
Here's the typical response of a behaved linear supply.  Forgive the appearance of the pic as it was quickly done.  The test load is the Array 3710a, and the PSUs are a $20 Radio Shack Micronta that has CV adjusts only and a Mastech 3030D, $70.

Array shows 1.50 A draw and 4.89V at the terminals.  The difference in voltage from the PSU are due to V drop on the cables.

The Micronta is set to 5.081V as shown on the Fluke and supplies 0.5A shown on its paltry analog meter  :-\.  How much current it provides in the share can be adjusted by setting the Mastech output voltage, in this case the Mastech supply's 1.0A.

The Mastech supply shows 1.04A at 6.0V and, set higher than the Micronta, provides most of the current.  To share the current equally between 2 supplies, adjust the higher voltage set supply lower, until the desired value.

You may not be able to set the supplies exactly the same by their onboard readouts unless you use remote sensing, because the V drop on the cables are also taken into account by the circuitry.  As cables and units heat up, the setting may drift slightly, so in the end, a stable balance can be had for something like 55:45 once your PSU are warmed up and in a stable state.

Finally, in balance both units are in CV.  When one PSU provides most of the current and the other PSU providing only excess to the load, the unit providing most of the current goes into CC mode.

The image is huge is show the readouts as true.

« Last Edit: June 16, 2014, 01:28:53 pm by saturation »
Best Wishes,

 Saturation
 

Offline thmjpr

  • Regular Contributor
  • *
  • Posts: 175
  • Country: ca
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #613 on: June 17, 2014, 05:22:19 am »
Thanks thmjpr for the detailed reply.  The basic response of a CVCC supply in parallel is well described; here's a good reference for the details:

http://www.home.agilent.com/agilent/editorial.jspx?ckey=520808&id=520808&nid=-11143.0.00&lc=eng&cc=US

http://powersupplyblog.tm.agilent.com/2012/11/how-can-i-get-more-power-from-my-power.html

When you adjust one channel's voltage higher than the lower voltage set channel do you notice it switch from CV to CC annunciator on the LCD screen?

Ok the link/pics helps me understand better what you were previously describing.
When I adjust the voltage higher than the lower voltage channel, it switches from CV to UR to CC over about 6mV range. Basically as you expected (with the added UR state).

Description of UR:
Quote
DP800A series power supply provides three output modes: constant voltage output (CV), constant current output (CC) and critical mode (UR). In CV mode, the output voltage equals the voltage setting value and the output current is determined by the load; in CC mode, the output current equals the current setting value and the output voltage is determined by the load; UR is the critical mode between CV and CC.
 

Offline mcinque

  • Supporter
  • ****
  • Posts: 1129
  • Country: it
  • I know that I know nothing
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #614 on: June 21, 2014, 01:31:06 pm »
My unit has been replaced last week after more than 8 months of requests.
 

Offline stormbr

  • Regular Contributor
  • *
  • Posts: 50
  • Country: br
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #615 on: June 26, 2014, 06:30:56 pm »
I receive yesterday my second unit of dp832, with the same problem of my first device.

Channel 2 and 3 locked on UR mode (critical mode). Or I'm a noob that not know adjust voltage on a power supply ?  :palm:


How to do for change the operation mode of channels from UR to CV/CC mode ?

The Rigol and Tequipment doing me paid again all taxes and shipment cost to brazil.

I spending 550,00 dolars yesterday just to received my "new" unit.

 |O





« Last Edit: June 26, 2014, 06:37:14 pm by stormbr »
 

Offline saturation

  • Super Contributor
  • ***
  • Posts: 4787
  • Country: us
  • Doveryai, no proveryai
    • NIST
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #616 on: June 27, 2014, 08:25:23 pm »
I do not own this unit.

I think UR stands for "unregulated", that is the third state of most any CV-CC type power supply.  Its often not mentioned because its "critical" that the supply not stay in this mode for long, or it will damage itself or the device its powering.

With the current at maximum, the leads open circuit, the channel must be in CV mode.  If you short circuit the outputs, it should jump to constant current mode.  It should not be in UR for any length of time.  PSU may go into UR mode when you turn the volt or amp adjust knobs for the most briefest of instances.  UR can also appear for a indefinite period once you sink current into a PSU that was not designed to sink current.

Best Wishes,

 Saturation
 

Offline centon1

  • Supporter
  • ****
  • Posts: 30
  • Country: ca
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #617 on: June 27, 2014, 11:11:36 pm »
How to do for change the operation mode of channels from UR to CV/CC mode ?

When all else fails, read the directions.

DP832 User guide dated Mar. 2013 covers it in section 2-1 'Front Panel Operations'
 

Offline geekGee

  • Supporter
  • ****
  • Posts: 49
  • Country: bm
  • IT Veteran, EE Newbie
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #618 on: July 26, 2014, 04:05:51 am »
Buyers beware... they are still in the sales channel.

I just received a DP832A today purchased new from an authorized source in the US and it is the "Bad Design" revision.  I am especially frustrated as I asked pre-sales if the units they had in stock were the newer revisions and was assured this was the case.

It came with firmware 1.05 and the "Certificate of Calibration" is dated 13-Jul-2013 so not surprising it’s the old revision after reviewing this topic.

I purposely haven't mentioned the supplier as I haven't notified them yet and didn't want to potentially bad mouth them before they have had a chance to act on my complaint.  Mistakes happen and at this stage I am more interested in how they respond.
 

Offline AndersAnd

  • Frequent Contributor
  • **
  • Posts: 572
  • Country: dk
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #619 on: July 31, 2014, 10:27:20 pm »
I purposely haven't mentioned the supplier as I haven't notified them yet and didn't want to potentially bad mouth them before they have had a chance to act on my complaint.  Mistakes happen and at this stage I am more interested in how they respond.
Have the replied yet?
 

Offline geekGee

  • Supporter
  • ****
  • Posts: 49
  • Country: bm
  • IT Veteran, EE Newbie
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #620 on: August 03, 2014, 03:33:19 pm »
I purposely haven't mentioned the supplier as I haven't notified them yet and didn't want to potentially bad mouth them before they have had a chance to act on my complaint.  Mistakes happen and at this stage I am more interested in how they respond.
Have the replied yet?

Yes… the supplier contacted Rigol then responded that the bad board was replaced earlier this year.  As I was almost 100% positive that was incorrect, I then contacted Rigol Tech Support directly.  Based on my description of the board from looking through the cooling holes, the engineer confirmed it was the bad board and has issued an RMA on the unit.

Rigol feel that it was likely their mistake.  As they are taking ownership of the issue, I am optimistic that I will have a positive outcome.
 

Offline stormbr

  • Regular Contributor
  • *
  • Posts: 50
  • Country: br
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #621 on: August 06, 2014, 03:13:07 pm »
Rigol is not serious company, low quality products and have no warranty.

3 years of warranty is a lie.  :--


I spending 4k $ with this company, but this was last time.


Rigol and Tequipment never more to me !

Two DP832 and one DG4062 damaged here.  |O


I thinking that happens only in Brazil ! Is not easy for anyone.
 

Offline f1rmb

  • Regular Contributor
  • *
  • Posts: 180
  • Country: fr
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #622 on: August 27, 2014, 06:13:05 pm »
Hi,

    This afternoon I was using my brand new DP832, powering a device under 24V (on channel 1), current was sets to 2A (but the device needs < 200mA).
Then, I had to leave the bench, so I turned off the channel (1), but when I came back, turning the channel back to ON, the power supply switched to 'UR', with few mV output.

I tried almost everything, turning off OCP and OVP, reboot, without any effect.

If I turn the channel ON, with few mV sets, then slowly increase voltage (using knob), the channel turns to 'UR' for a sec or so, then switch back to Constant Voltage. But if I increase too fast, it sticks to 'UR'.
Also, setting output voltage to ~1V, then turning channel output ON seems to almost work (voltage slowly increase, but doesn't reach 'UR' state).

Does anyone experienced the same problem, or have any clue or fix ?

The firmware version is 1.10, all options "turned" ON.

Cheers.
---
Daniel.
« Last Edit: August 27, 2014, 06:14:48 pm by f1rmb »
 

Offline mcinque

  • Supporter
  • ****
  • Posts: 1129
  • Country: it
  • I know that I know nothing
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #623 on: August 28, 2014, 12:50:21 pm »
TFM (that I guess you have read  :P) says:

"In CV mode, the output voltage equals the voltage setting value and the output current is determined by the load; in CC mode, the output current equals the current setting value and the output voltage is determined by the load; UR is the critical mode between CV and CC."

I think this behaviour is caused by the load. You should try to power another device to see if it behaves the same.
 

Offline f1rmb

  • Regular Contributor
  • *
  • Posts: 180
  • Country: fr
Re: EEVblog #512 - Rigol DP832 Bad Design Investigation
« Reply #624 on: August 28, 2014, 05:22:20 pm »
Hi,

    Okay, problem found and fixed.
My reseller asked me to ship back to him the device. Noooo way I spend half the price of this baby in UPS/DHL/whatever shipment.

The positive terminal on the top board has so badly soldered that I pulled it out when I tried to remove the wires.

Also, one of the soldered fuse (5A) on the same board was almost dead (this explains why the voltage was increasing so slowly.

For test purpose, I temporary straped the fuse, and the channel is working again. Now I need to find a replacement.
« Last Edit: August 28, 2014, 08:31:05 pm by f1rmb »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf