EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: IamSynthetiC on June 08, 2020, 12:51:22 pm
-
Greetings and i hope you are all well.
Lately a lot of projects require some testing of power supplies , batteries , converters and such.The LM324 electronic load i have, offers only CC, not CV,CP or CR.
Therefore i would like to either improve upon it or buy one.
DIY.
I have "half designed" a solution for the load which uses an arduino and external ADC and DAC. I plan to use the DAC to output a set voltage onto the feedback of the LM324 (meaning, to replace the potentiometer with the DAC), and add the CV,CP and CR functionalities in software.
The problem lies in the limited time i can spend on this, since this is but a hobby of mine, while i am studying physics (and finals are coming up), and i work part time.
BUY
I have been looking at the Rigol DL3021 and the Siglent SDL-2020X-E for some time now. I am willing to pay the price for such equipment but before i do, i would like to get your opinions. Also second hand equipment is an option.
To conclude:
- Are there any alternatives to the models i have mentioned?
- Are there any good models i can search to find second hand?
- Is eBay even worth looking at (customs can be expensive)?
Thank you for your time.
I am looking forward to hearing your thoughts.
Have a good day, and take care.
-
Look at the KEL-102 and KEL-103 models. Those seem to provide excellent value for money. Other deals can be had but you need to get lucky. DC loads aren't that commonly sold second hand.
-
I would recommend taking a look at the Kunkin KP184, I've done a series of videos on this model and it's pretty good for it's price. You do need to be careful about where you buy it from because the latest revision seems to have fixed a few issues and is preferred.
https://www.youtube.com/watch?v=mPRSRuvg3M8&t=542s (https://www.youtube.com/watch?v=mPRSRuvg3M8&t=542s)
-
Look at the KEL-102 and KEL-103 models. Those seem to provide excellent value for money. Other deals can be had but you need to get lucky. DC loads aren't that commonly sold second hand.
+1
Have one, exluding the horrible user interface (you need to have always the user manual under the hands), the voltages are pretty accurate, can work quite well on really low loads (1ma) and have a lot of functions).
-
Rigol 3021
Buy one if budget supports it and focus on the problem to be solved vs designing something that already exists.
-
Rigol 3021
Buy one if budget supports it and focus on the problem to be solved vs designing something that already exists.
I don't think Dave like that one very much.
-
Thank you all for your responses.
i have considered the KEL linup, aswell as the Kunkin, for a long time, as they seem to be the Go-To option when it comes to low-end cheap electronic loads.
However the load i am designing myself will unfortunatelly probably end up being a low-end electronic load.
I received some parts yesterday and i spent the day today testing them. They should allow me to control the electronic load i have, with a satisfactory accuracy. There is a long way to go still, but its coming together.
So, to add to my previous question, what is it that i gain from a more expensive electronic load ? What makes and electronic load "High-Quality"? Is it features, quality or something else ?
And more importantly is it that hard to replicate them ?
I feel it is important to say that by no means do i wish to undermine the people thet build and design these. I just feel like with all those teardown Dave and other people have done, some of the "secrets" would be spilled.
And with so many people ,on youtube and here, building their own, why does everyone build the same exact design ? Why is that "Low-end" ?
I am afraid i am going to get slapped in the face by the Dunning-Kruger effect. But I'm ready to learn !!!
Also, what do you guys think of the ITECH IT8511+. Primarily comes from china but seems to have everything, including dataloging which i really enjoy. Haven't found a teardown or a review (apart from Andreas Spiess), what does that mean ?
Thank you again, and i am eager to hear your thoughts.
-
So, to add to my previous question, what is it that i gain from a more expensive electronic load ? What makes and electronic load "High-Quality"? Is it features, quality or something else ?
And more importantly is it that hard to replicate them ?
Compared to a low end DIY type load - and generally speaking...
High(er) accuracy, high(er) resolution, adjustable rise/fall times, CC/CV/CP/CR modes, battery discharge testing, pulsing, ramping, sequencing, logging and of course remote interfaces like LAN, GPIB, RS232, USB. Perhaps it comes with some more or less usable/polished PC software.
Now, if and more importantly how well, these features are implemented on the various loads you can buy will certainly differ (I suppose that falls in the quality category) so you'd need to decide what's important and read up and compare. How hard these are to implement on a DIY device obviously depends on your level of expertise.
I have a Maynuo unit myself. It (and the frequency counter) is by far the least used unit in my home lab but whenever I DO use it it does what I expect from it. Recently there's been some discussion on how well these perform at very low current levels. If that's important to you you may want to search for that thread. IIRC the Rigol DL-series was especially bad at that.
-
If you need CC accuracy and stability in the mA region, be aware of the limitations of some lower cost models. As an example, My Tekpower 3710A provides a CC of 0.832 mA when set for 0 mA or 1 mA. Also, there is quite a bit of hysteresis in the 0 to 5 mA region. Instabilities also exist: when set to 10 mA the current as measured by a calibrated Fluke 289 varies from 8.54 mA to 8.64 mA. Finally, the available software is very marginal.
The above is not complaining. It performs as well as expected. It all depends on needs and budgets.
BTW, I am looking for battery testing software for mine.
-
I hadnt considered the complexity of the circuit as much, until i searched around and found this : http://www.ti.com/solution/dc-power-supply-ac-source-electronic-load. (http://www.ti.com/solution/dc-power-supply-ac-source-electronic-load.) Which is complex allright.
Also, ofcourse with higher price, comes (usually) greater quality, more features and such. The thing that proplexes me though is the fact that the market is very stale at the lower ranges. There seems to be ,either no competition, or demand (or even interest).
I mentioned before the absence of high-quallity (<---high quality is relative here) DIY alternatives. Which is untrue, as Scullcom has made an excellent version. I have watched all the episodes (6-7 hours) but my brain YEEETED it all out. :palm:
The fact remains though, that for a hobbyist, or an amature, paying 400+ for what ends up being something extremely basic in its category, is not pretty looking option. The standard features accross the range for such devices seem to only be the load's own functionality and the meters.
To come accross features such as these bellow, you'll already be searching at the 550+ range
Compared to a low end DIY type load - and generally speaking...
High(er) accuracy, high(er) resolution, adjustable rise/fall times, CC/CV/CP/CR modes, battery discharge testing, pulsing, ramping, sequencing, logging and of course remote interfaces like LAN, GPIB, RS232, USB. Perhaps it comes with some more or less usable/polished PC software.
As for my project, i think i will go with the DIY route during the summer and see where it leads me, simply out of spite 8). I have already managed to implement a INA226 Voltage-Current-Power sensor on the arduino that monitors the load is satisfactory intervals. I now only need to have it control it to get the basic functionality out of it.
Grnadchuck : There was a thread the other day regarding software someone was designing for PC control to varius devices, and he claimed it was fairly easy to incorporate new devices. Unfortunately i can seem to find it. Perhaps someone elses can.
-
Since you were prepared to do DIY and I don't know what specifications you were after - I thought it might help if I 'chime-in' with a cheap - 'get you by for now' solution ...
I have one of these devices - it works suprisingly well and is true to its specifications ...
https://www.amazon.co.uk/Constant-Current-Electronic-Discharge-Capacity/dp/B07WVC2FS8/ref=sr_1_1?dchild=1&keywords=electronic+load+180w&qid=1591814947&sr=8-1 (https://www.amazon.co.uk/Constant-Current-Electronic-Discharge-Capacity/dp/B07WVC2FS8/ref=sr_1_1?dchild=1&keywords=electronic+load+180w&qid=1591814947&sr=8-1)
Its about $50 - nothing copared to a pro device both in price and specifications / performance
BUT
It may work for your immediate needs :P
-
I've just received a Kunkin KP184 and it seems to do what it says.
Couple of (serious) problems though:
1) Calibration seems off under 50mA.Probably need to recalibrate it even though I just got it out of the box. :palm:
It seems far more accurate after 60-70mA so depending on your needs this might be a deal breaker.
For me it's certainly a problem.
This might just be my unit though; I've seen some YT videos prior to buying and it seems that people are generally happy with this.
2) The fan is always on. (and it's noisy!)
Again, I may have received a faulty/defective unit but the fan on mine is constantly at full speed it seems.
I consider it very noisy for a lab environment.
EDIT: Fan control was problematic as the transistor near the fan connector was barely touching the PCB, only half of its three legs was attached.
I resoldered it back in its place and it's now OK.
Fan is now very quiet even under moderate stress.
So the fan noise is not an issue with these if they work properly..
But,
My unit wouldn't pass the calibration test or even a simple inspection test with that transistor missing.
The problem it was having was very noticeable by almost anyone.
So my guess is that the calibration leaflet they provide with the unit is useless.
They probably store some generic calibration values on the MCU (?) and ship it off when it's assembled.
Given my case I seriously doubt even a simple power-on test is performed after these are built.
So if anyone is looking for a KP184 be prepared to recalibrate yourself if you need it to be in spec.
-
If your power requirements aren't that high and are interested in more precision, you might want to consider one of the older HP dynamic measurement sources 66332A or 66312A.
These can often be found for cheap as they were used extensively in the early days of cell phone testing and have now been phased out for newer SMUs.
They can do a lot more than an electronic load, e.g. four quadrant operation for things like battery simulation, but are more limited in power levels and slew rate.
-
So, to add to my previous question, what is it that i gain from a more expensive electronic load ? What makes and electronic load "High-Quality"? Is it features, quality or something else ?
And more importantly is it that hard to replicate them ?
You gain time actually using it IMO.
If you need it for a specific purpose you may benefit more from a ready made solution instead of a DIY one.
Even if the actual design is ready for you to implement all the details like the case, wiring, heatsink, transfomer, connectors, LCD or some kind of interface etc quickly add up in time spent researching and of course cost.
And all that with the fear that you will probably end up with a "DIY looking" piece even if it works fine.
There are quite a few designs hanging around even advanced ones and I've been thinking to made one myself but as I said above I just bought the cheapest unit that actually looks sturdy enough for lab usage.
I don't know your needs but my own opinion is that I don't want to DIY test gear.
Especially if it's involving power electronics.
Tried it in the past and I ended up with weird looking boxes that felt mediocre to use.
Just my opinion though :-+
-
Thank you all for your interest and your thoughs on this.
As far as my current needs are concered, i have managed to get a friend ,of a friend ,of a friend ....... to lend me his Siglent SDL-1000X (What a kind man)
I'll certainly be able to proceed with my projects now.
I have one of these devices - it works suprisingly well and is true to its specifications ...
https://www.amazon.co.uk/Constant-Current-Electronic-Discharge-Capacity/dp/B07WVC2FS8/ref=sr_1_1?dchild=1&keywords=electronic+load+180w&qid=1591814947&sr=8-1 (https://www.amazon.co.uk/Constant-Current-Electronic-Discharge-Capacity/dp/B07WVC2FS8/ref=sr_1_1?dchild=1&keywords=electronic+load+180w&qid=1591814947&sr=8-1)
I originally considered this as a serious alternative to an electronic load, but i then saw its smaller brother (an 10$ LM324 basic Kit) and ultimately decided that i wanted to DIY it.(People often create their own problems :D)
As for the the Kunkin KP184,i always regarded it as a cheap tool, however i've seen it sitting in far more professional benches than mine. Therefore it must be doing something right. Also it is very hard to argue with that price.
You gain time actually using it IMO.
If you need it for a specific purpose you may benefit more from a ready made solution instead of a DIY one.
One can argue that if specifics are required, one can make it themselves ,to be as specific they need. The market is flooded with all-around-tools most of the time. (lots of general things mentioned here, so not going to dive deep).
The thing i want to mention is the time. Our efforts will probably never be as good as the collective effort of experts, with long-time experience in this field (access to resources, materials, tools .....).
But for me ,and certainly for many people more, it is the satisfaction of seeing your ideas and efforts take shape, and making something of your own.
Which ultimately takes a lot of your time. (to build the device, find its limits ,and to build trust in it). So the question each has to answer is "is my time worth it ?", in with i agree ,most of the time, the answer is "No".
If your power requirements aren't that high and are interested in more precision, you might want to consider one of the older HP dynamic measurement sources 66332A or 66312A.
KrudyZ, you gave me an epiphany. I never even though of an SMU before, due to their more than vast pricetag. These ones you mentioned certainly dont fit all my needs for a load, but i've been needing a second power supply aswell as a small load. I will certainly look into it. Thank you
-
Hi,
I am looking for a decent DC electronic load (150V/30A/300W) and come across with some Chinese models I have seen no reviews so far.
Tonghui
http://www.tonghui.com.cn/en/product_956.html (http://www.tonghui.com.cn/en/product_956.html)
Hopetech
https://www.hopeinstrument.com/product/programmable-dc-electronic-load-150w-300w-series/ (https://www.hopeinstrument.com/product/programmable-dc-electronic-load-150w-300w-series/)
Uni-T
https://www.uni-trend.com.cn/index.php?m=content&c=index&a=show&catid=475&id=835 (https://www.uni-trend.com.cn/index.php?m=content&c=index&a=show&catid=475&id=835)
The first two models have 500KHz sampling rate, while the last one has 40KHz.
Someone heard about them or owns one?
-
What kind of voltage and power dissipation do you need?
-
Have a look at the BEICH CH9720 series: https://www.aliexpress.com/item/10000109988978.html (https://www.aliexpress.com/item/10000109988978.html)
I'm quite happy with mine, and I found the seller quite good.
-
Grnadchuck : There was a thread the other day regarding software someone was designing for PC control to varius devices, and he claimed it was fairly easy to incorporate new devices. Unfortunately i can seem to find it. Perhaps someone elses can.
That is probably my software TestController: https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/ (https://www.eevblog.com/forum/testgear/program-that-can-log-from-many-multimeters/)
It is fairly easy to add SCPI devices, devices with a ascii based protocol, MODBUS and multimeters (serial type), most other stuff may require working together with me.
It do support over 300 different devices at the current time, mostly: multimeters, power supplies, electronic loads, but there are other stuff. All the different devices can be logged together and usually also controlled.
-
What kind of voltage and power dissipation do you need?
I think up to 150V and 300W are ok for me.
-
Have a look at the BEICH CH9720 series: https://www.aliexpress.com/item/10000109988978.html (https://www.aliexpress.com/item/10000109988978.html)
I'm quite happy with mine, and I found the seller quite good.
It is an interesting model. I will study a bit more the specifications.
Have you checked the minimum current rise and fall times in Transient test (say from 0A to 3A/30A)?,
Have you checked the current setting accuracy at low values (say 3mA in 3A range) ?
Have you found stability problems in CV mode?
-
It is an interesting model. I will study a bit more the specifications.
Have you checked the minimum current rise and fall times in Transient test (say from 0A to 3A/30A)?,
Have you checked the current setting accuracy at low values (say 3mA in 3A range) ?
Have you found stability problems in CV mode?
Andrusca - under all conditions I have needed, I have found the accuracy to be superb. I primarily use it to test very small cell discharge characteristics for my customers. I have not fully characterized all modes.
As I don't have a lot of spare time (#$!% customers!! :P) - if there is a specific test case you would like me to look at, give me full connection details and I'll attempt to find the time to evaluate.
Regards,
Torsten
-
I checked the BEICH homepage and only found a Chinese manual and no SCPI documentation.
-
HKJ - I received a CD with an English manual. The quality of the English is quite good.
The manual introduces the remote control of the equipment, but doesn't provide full detail.
I suspect a full protocol manual exists as the manual mentions the following:
"Only the use of interface is introduced in this chapter, please refer to CH9720 program protocol for command."
-
do you know the difference between the B and BU models?
-
The "U" versions have U-DISK (USB) - for SW updates and saving.
-
sorry, it's still not clear to me, is it: without the U they don't have the internal storage or the usb port in the front? or with it: they have an internal storage device besides the usb port?
-
From the AliExpress page:
"■ Software update of U disk, data saving and screen capture function of U disk (CH9720BU/9720CU)."
I have a "U" version. It has USB.
Therefore:
CH9720B & CH9720C = no USB
CH9720BU & CH9720CU = USB
-
Hi, CDN_Torsten
That battery test has convinced me about the DC load accuracy in steady state at low current levels. Very nice!
Before buying a DC load I would really love to check its performance under these two other applications:
1) Current step response of DC/DC power supplies:
Here I would set CC TRANSIENT mode and I would perform one pulse
a) 0A -> 20A and then 20A -> 0A with V(DC)=15V (if the power limit is an issue, I would use V(DC)=7.5V)
b) 0A -> 1A and then 1A -> 0A with V(DC)=300V (if the power limit is an issue, I would use a current step of 0.5A instead of 1A)
In both situations, the high and low level time would be 500us.
I would like to see which are the minimum current rise and fall times resulting of these experiments.
2) Resistance emulation in single-phase AC/DC rectifiers:
I would connect the DC load to the output of a standard bridge rectifier sourced from an AC variac transformer. I would then set the electronic DC load in CR mode to emulate a resistor connected to a full wave line-rectified voltage and test how accurately the current can follow the voltage variations. I would adjust the peak rectified voltage V(pk)=300V and would set the emulated resistor R=300ohm. If there is a power limitation, I would use R=600ohm. The current measurement range is set to 3A.
Hey, all that is certainly too much to ask you to test, but perhaps you could arrange something of the sort according to your possibilities to give me an idea of the DC load transient performance. However, while I am writing this I am seriously thinking about kindly asking BEICH to do those experiments for me. Maybe they are really eager to sell their stuff and would not mind to please a customer…
-
Andrusca - I am unable to run test#1 at the current level you have requested.
The best I can do is 10A. I also don't know specifically what you are looking for...so here is a kick at the can:
Setup: Power supply (15V) -> 37000uF capacitor -> current clamp -> Beich Load
Load Setup: 0A - 10A (continuous cycling), 0.6A/us rise and fall time, time at each current level of 0.5ms.
Attached is the current waveform. Is this what you are looking for??
Regards,
Torsten
-
Here is the same test with a slower edge (10mA/us).
-
Andrusca - I am unable to run test#1 at the current level you have requested.
The best I can do is 10A. I also don't know specifically what you are looking for...so here is a kick at the can:
Setup: Power supply (15V) -> 37000uF capacitor -> current clamp -> Beich Load
Load Setup: 0A - 10A (continuous cycling), 0.6A/us rise and fall time, time at each current level of 0.5ms.
Attached is the current waveform. Is this what you are looking for??
Regards,
Torsten
Hi Torsten,
Yes, this is basically what I wanted to see. Thank you very much. It seems that the best obtainable current slew rate is about 0.12 A/us. This is almost an order of magnitude below what I would want for testing a power supply transient response. Here we are assuming that the slew rate of the current clamp and the current capacity of the power supply you are using are not limiting the measurement.
Regarding test #2, I am just exploring the possibility of transforming the DC load into an AC load (an adjustable resistor, in fact) by placing it inside a full-wave bridge rectifier.
All these use cases are useful for me to test my power converter designs.
Regards,
Andrés
-
Hi Andrés,
I'm assuming that you are basing the best slew rate on the R/C-like curve of image #1. I would recommend that this needs further investigation as you may be (are) seeing the slew rate of the current-probe which is only rated to 20kHz. A better setup would be to measure the voltage developed across a resistor with a scope. I just didn't have one capable of surviving the 10A current.
Regards,
Torsten
-
Hi,
I'm interested on Beich dc load but I'm unable to find a manual to understand the functions of the device.
Anyone has found it?
Can it simulate a diode V/I?
Regards
-
AcM - here is the manual.
-
Each part has all the material in PDF read the description.
https://youtu.be/0nS7nirg_xM
https://youtu.be/cczs_6Mu2-Y