Author Topic: EEVblog #315 - Korad KA3005P Review/FAIL  (Read 251887 times)

0 Members and 1 Guest are viewing this topic.

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
EEVblog #315 - Korad KA3005P Review/FAIL
« on: July 18, 2012, 04:20:08 am »


Dave.
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #1 on: July 18, 2012, 04:33:48 am »
I know that beeper sound, it's using a beeper that requires the micro to generate the tone. (They're cheaper than beeper with internal tone generators)

When you try and generate the ~2kHz by setting/clearing the mcu output pin in software other interrupts will semi-randomly delay the set/clear commands and you get a sound that's dirty like that.

To avoid it they need to use the dedicated hardware timer output pin on the mcu so the tone generation happens independently.
(or a beeper that does the tone generation itself)

« Last Edit: July 18, 2012, 04:39:34 am by Psi »
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline snoopen

  • Contributor
  • Posts: 28
  • Country: au
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #2 on: July 18, 2012, 05:36:12 am »
Wondering if you'd recommend this over the similarly priced Jaycar lab supply? Looks like they both have an overshoot issue.

This is the one:
http://www.eevblog.com/2009/09/11/eevblog-30-jaycar-bench-lab-power-supply-review/

I recently bought the Jaycar supply, but it's unopened in the box... Thinking I could take it back and get the Korad for the extra $40 or so.
 

Offline snoopen

  • Contributor
  • Posts: 28
  • Country: au
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #3 on: July 18, 2012, 05:41:44 am »
Hmmm on second thoughts, I actually finished watching the last 5 mins of the review. Can see why you say FAIL! :O Oh wells!
 

Offline Erwin Ried

  • Regular Contributor
  • *
  • Posts: 201
  • Country: no
My website: http://ried.cl
 

Offline nitro2k01

  • Frequent Contributor
  • **
  • Posts: 843
  • Country: 00
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #5 on: July 18, 2012, 05:55:08 am »
Playing devil's advocate here, but could it be that you've swapped the power input and output on the logic board when reassembling it after the teardown? They seem to be using the same connector.
This way you  might for example have blown the ever so beefy protection diode across the shunt, and later voltage regulation transistor or something else.
« Last Edit: July 18, 2012, 05:57:57 am by nitro2k01 »
Whoa! How the hell did Dave know that Bob is my uncle? Amazing!
 

Offline dda

  • Contributor
  • Posts: 46
  • Country: au
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #6 on: July 18, 2012, 06:14:49 am »
Dave did get into M5 at one point in the vid, but he didnt make any note of it
 

Offline Psi

  • Super Contributor
  • ***
  • Posts: 9889
  • Country: nz
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #7 on: July 18, 2012, 06:18:07 am »
Dave did get into M5 at one point in the vid, but he didnt make any note of it

"M5" is Mission Impossible mode for MI5 operatives.

This unit will self destruct in 5 seconds.
Greek letter 'Psi' (not Pounds per Square Inch)
 

Offline dcel

  • Regular Contributor
  • *
  • Posts: 179
  • Country: us
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #8 on: July 18, 2012, 07:37:11 am »
Dave

An autopsy would be nice so we can see what failed and showcase your troubleshooting skills for all the young players.

Chris
 

Offline hans

  • Super Contributor
  • ***
  • Posts: 1626
  • Country: nl
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #9 on: July 18, 2012, 08:28:44 am »
Honestly, a follow-up would be funny. It would humiliate this PSU even more.

I wasn't impressed with the teardown stage. Everything is 'there' to get a DC programmable PSU, but it seems rubbish built. Uncleaned PCB's, random holes drilled, poor soldering, weird R2R DAC (apparently PCB space and SMT parts placing is free for them) and overall shitty stuff :)

That you have killed it with a 'constant resistance' is too funny. That's like connecting 1 ohm on a 24V output, you should get 5V @ 5A right..

I think the very odd load transients are maybe due to it's unusual current shunt. Seen it in the teardown? It's like more like an inductor of several tenths to hundreds of uH. I would be interested in setting the load transient at turn-off (at least as important as turn-on), will it overshoot due to the inductor?
But unfortunately it's died already :-\

« Last Edit: July 18, 2012, 08:30:42 am by hans »
 

Online David Aurora

  • Frequent Contributor
  • **
  • Posts: 413
  • Country: au
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #10 on: July 18, 2012, 08:30:36 am »
My curiosity is killing me here haha, I'm dying to know more about the death of this thing. Did any magic smoke actually escape, or is it software related?
 

Offline Physics_Dude

  • Newbie
  • Posts: 2
  • Country: 00
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #11 on: July 18, 2012, 09:00:22 am »
Something like this happened to me a while ago with a basic 30v/8A switched mode DC power supply. I think it was a Mastech brand, but I could be wrong. I was using it to power an innocent little project at 6 volts with only a few milliamps, when suddenly at random, the power supply decided to dump a steady 56 volts through the little thing and kill it! :(
Further testing relieved that the power supply was also in a consent current mode of around 8 amps. None of the output adjustment knobs had any effect on the output voltage or current.

It was replaced soon after by a more preferred power supply, but I decided kept the broken one, it seemed quite reliable for supplying the 55-56 volts that it was stuck at.
However, after weeks of abuse on the old one, while I was using it under a fairly heavy load for some extend period of time (the device lacks a cooling fan and will get hot), I heard a few clicks of the relays, and wouldn't you know it, it went back to its normal, advertised operation. I had my full voltage and current limiting controls back too!

I don't use this zombie-like power supply for any serous projects anymore for obvious reasons. It also recently decided to get stuck again in consent current mode at 8 amps. I still have full control over its output voltage though.

I only mention this because Dave seems to have fallen in a similar predicament where his power supply seems to send out its full current capacity at a higher then rated voltage (42.3v at first).
 

Offline Scopeman1

  • Contributor
  • Posts: 11
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #12 on: July 18, 2012, 09:11:03 am »
Just been to TRIO's website.  How well do you know these guys Dave?  Are they any good?  Agilent all over the place.

I saw this on link below......    does not look too bad for the money, and they are local in Sydney.  It's better value than the Jaycar.  It is a triple output for less than the Jaycar dual output cost even with GST & freight!!

http://triosmartcal.com.au/dc-bench-supplies/2432-atten-aps3003s-3d-dual-power-supply-0-30v-3a.html

They have a 150W basic bench supply for $99  http://triosmartcal.com.au/dc-bench-supplies/2631-qje-ps3005-bench-power-supply-150w-0-30v-0-5a.html   It can't be a linear for that price.

As for cheapie scopes, I see they are selling the same scope as Jaycar for almost half the price as Jaycar !!!  WTF!!   They can't be ripping people off at that price or  .... Mr. Jaycar is charging us too much for convenience or ????

http://www.jaycar.com.au/productView.asp?ID=QC1932  jaycar $599 inc GST
http://triosmartcal.com.au/2195-siglent-digital-storage-oscilloscope-25-mhz-bandwith-4-channel.html  = TRIO $298 ex-GST

It's the same scope isn't it? Can you confirm?
 

Offline wiedehopf

  • Newbie
  • Posts: 2
  • Country: de
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #13 on: July 18, 2012, 09:30:51 am »
Playing devil's advocate here, but could it be that you've swapped the power input and output on the logic board when reassembling it after the teardown? They seem to be using the same connector.
This way you  might for example have blown the ever so beefy protection diode across the shunt, and later voltage regulation transistor or something else.

I think the power cable leading from the control board to the output ist to short to connect it to the input on the control board. Dave would've noticed the cable being tense, so it's not an error i think he would've made.
But i guess we'll see soon enough when he tracks down the error.
 

Offline EEVblogTopic starter

  • Administrator
  • *****
  • Posts: 37661
  • Country: au
    • EEVblog
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #14 on: July 18, 2012, 09:35:33 am »
UPDATE: Triosmartcal have withdrawn it from sale, and Korad are investigating...

Dave.
 

Offline PChi

  • Frequent Contributor
  • **
  • Posts: 264
  • Country: gb
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #15 on: July 18, 2012, 10:22:04 am »
Thanks for the video tear down and test.

It might have foldback on the current limit which may be why it was giving inconsistent results with the constant power load.

I am guessing that the output on/off switch just controls the Power Transisters which may have failed Collector Emitter short circuit. The output stage might rely on never having to drive a short circuit when the transformer tap is at the maximum voltage for survival (even with some form of foldback limiting). Possibly Korad need to check the Safe Operating Area of the output transistors. Also cheap power transistors can be unreliable.
 

Offline amyk

  • Super Contributor
  • ***
  • Posts: 8240
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #16 on: July 18, 2012, 10:24:48 am »
I would've preferred a relay-switched output. Those usually do not fail short (unless the contacts have overloaded and welded together.)
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3697
  • Country: sg
    • T4P
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #17 on: July 18, 2012, 12:30:52 pm »
I would've preferred a relay-switched output. Those usually do not fail short (unless the contacts have overloaded and welded together.)

The golden method of switching voltages in and out without a physical switch but obviously... dickheads-designers
and yes a crowbar is better
« Last Edit: July 19, 2012, 05:13:22 am by DaveXRT »
 

Offline Greg J

  • Regular Contributor
  • *
  • Posts: 83
  • Country: gb
  • Hi there
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #18 on: July 18, 2012, 01:18:59 pm »
Chinese quality. I don't think anyone was hoping this will work for this price.
Even their website is one big fail.
« Last Edit: July 18, 2012, 01:21:02 pm by Greg J »
--
Take It EV
 

Offline Stephen Hill

  • Regular Contributor
  • *
  • Posts: 178
  • Country: gb
  • M3VXY
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #19 on: July 18, 2012, 01:28:00 pm »
Even their website is one big fail.

What's wrong with their website?
http://koradtechnology.com/en/
 

Offline Bloch

  • Supporter
  • ****
  • Posts: 453
  • Country: dk
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #20 on: July 18, 2012, 02:26:01 pm »
 

Offline Rufus

  • Super Contributor
  • ***
  • Posts: 2095
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #21 on: July 18, 2012, 02:26:14 pm »
It is pretty obvious the main series regulating transistor failed short circuit providing an unregulated supply from whatever the transformer tap relays have selected.

When you start putting effective shorts on the output with the supply set to high voltage the pass transistor experiences maximum stress until the relays tap down the transformer. Probably an SOA failure because the output transistor is grotty or marginally specified.

On 'funny things going on' understand a constant power load with no volts applied is trying to be a short circuit so when you switch a supply on to it the supply ought to go straight into current limit and stay there delivering insufficient power.

The load probably has a minimum turn on voltage setting/option where it won't start drawing current till the voltage reaches that value as a workaround for the problem.

When you turn on a constant power load with a supply already connected it is a race between the load reducing the current draw and the supply going into current limit.
 

Offline slburris

  • Frequent Contributor
  • **
  • Posts: 542
  • Country: us
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #22 on: July 18, 2012, 02:49:19 pm »
I almost wonder if they are doing something stupid, like implementing the control
loop in the processor, given the strange ramps at turn on.

Of course you would get terrible transient response doing it that way, but
I've seen designs on the net doing just that.

Scott
 

Offline firewalker

  • Super Contributor
  • ***
  • Posts: 2450
  • Country: gr
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #23 on: July 18, 2012, 03:15:52 pm »
The machines that checks if a chip passes all the tests what does with the rejected ones? Maybe makes a mark on the surface? Just saying... You know, sanded chips...  :P :P :P

Alexander.
Become a realist, stay a dreamer.

 

Offline Electr0nicus

  • Regular Contributor
  • *
  • Posts: 131
  • Country: at
Re: EEVblog #315 - Korad KA3005P Review/FAIL
« Reply #24 on: July 18, 2012, 03:56:45 pm »
Most high current linear power supplies use more than one series transistor. They are usually paralleled and to eliminate thermal runaway there are current balancing resistors used. Maybe there are no, or insufficient current balance resistors used in this PSU, so one transistor gets most of the current and self destructs, so the other transistor also gets the maximum current and gets destroyed too.
« Last Edit: July 18, 2012, 03:59:14 pm by Electr0nicus »
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf