DC-Load wanted(Siglent SDL1020X-E found)

[Martin72]

Edit: Title changed
Hi folks,
I have almost everything a "private laboratory" needs, except for a spectrum analyzer and - loads...
I would like to change that first before I buy an SA.
First a DC load, I still have to think about AC, maybe I'll build it myself.
So DC load... Which one...
I would like to have one that I can regulate precisely to the mA, so that if I set a 5mA load, for example, it stays that way.
Then I should also be able to make load jumps with it to test control loops.
Furthermore, it should not interfere too much in order to be able to make reasonable ripple measurements.
Maximum current...Hm...Let's say 20A, more if you like.
Voltage 50V, also more if you like.
Price...Doesn't necessarily have to be the cheapest, but as cheap as possible under the conditions mentioned above, just the usual.
Make suggestions, I don't care about the brand.
Oh yes, and sensible connections would not be bad, preferably terminals where you can also connect 4mm bananas.

Thx in advance,
Martin

[blackdog]

Hi Martin,

It is difficult to find a "Dynamic Load" that can handle small currents and e.g. 40-Ampere properly.

It's a bit like with LAB power supplies you need several for certain applications.

Often you want to work with small currents and rapidly varying loads like microcontrollers and i.o.t devices.

So you want a good fast response, the dynamic loads that can handle large currents usually have an RC combination across the output terminals, because of this the peformance at smaller currents is not as good anymore since the low impedance vnadit RC network.

Maybe build something yourself where you can control some properties yourself?
For some inspiration take a look at this topic of mine on a Dutch forum
for a small fast active load.

https://www.circuitsonline.net/forum/view/121338/1

And a nother one
https://www.circuitsonline.net/forum/view/106770/1

And one special
https://www.circuitsonline.net/forum/view/153438

Have fun!

Kind regards,
Bram

[Martin72]

Hi Bram,

Very nice...Must diving deeper into it, thanks for the links.
The topic came up because I realized yesterday that I don't actually have any electrical loads here, except for something really bad from the China man.
In my test field we have umpteen loads from "Zentro", as well as two from Rigol, which a developer colleague once bought for a project.
I had a brief look at them, you can set the risetime there, which I found interesting, but the operation itself is terrible.
Then we now have loads from Statron, for DC and AC, whereby the AC load is pretty cool:
You do have a fine pot to slowly ramp up the load, but when you turn that, it clicks inside- It switches real resistors.
I think I'm going to have to go in two directions, one for larger flows and one for smaller and more dynamic flows.
As I said, I'll take a closer look at your design*, don't be surprised if I report on it in this thread, because that's when I started to rebuild it.

Martin

*(A few years ago I built myself a small load with several inputs because the test object had several outputs.
Quite simple with an OP amp and "afterburner".
The set currents always drifted away, so it ended up in the trash instead of thinking about why.)

[tszaboo]

Does the Keithley DC Electronic Load Series 2380 fit the budget? I have that at work, nice unit. Also tests batteries. For anything complicated, more than two current levels, you need to script it though.
I plan to buy a Riglol DL3000 at home, no experience with it yet. The base price is attractive.

[Martin72]

Hi,

Does the Keithley DC Electronic Load Series 2380 fit the budget?

3300€.....No.

I plan to buy a Riglol DL3000 at home, no experience with it yet. The base price is attractive.

Maybe it's also something for me, for larger loads.
As I mentioned, we have two of them at work, I'd have to take a closer look at how they work and if you want to know anything specific about them, I can test them.

[Grandchuck]

Hi Martin. 
The photo shows 10 minute drift anter 30 minutes warm up.

[Warhawk]

Martin, I went for Korad KEL102 from Reichelt. Yes, it is Chinesium but surprisingly reasonable quality.It is a good all-rounder. I like LAN and the standard LCD. Upgrade when you need to. DC loads are annoyingly expensive.

[Grandchuck]

After over an hour, still drifting up.

Photo looks better too!

[tszaboo]

Hi,

Does the Keithley DC Electronic Load Series 2380 fit the budget?

3300€.....No.

I plan to buy a Riglol DL3000 at home, no experience with it yet. The base price is attractive.

Maybe it's also something for me, for larger loads.
As I mentioned, we have two of them at work, I'd have to take a closer look at how they work and if you want to know anything specific about them, I can test them.

I think I know what you mean. There is a lack of ~32V, 10A capable DC loads, that would come with a front banana jack. Only C brands seem to do this. But writing this, I just remembered, Itech has DC loads, the IT8211 would fit your basic specs. I don't have any experience with the brand, but they come up a lot in discussions. This one, I don't see any digital interface on it.

[Veteran68]

Being somewhat of a Siglent fan, I'm surprised you've not mentioned the SDL series.

The SDL1020X-E can be easily "upgraded" to the SDL1030X with 0.1mV/mA resolution and 300W max power. 5A/3A and 36V/150V ranges, configurable slew rates, battery test mode...

Also the binding post knobs are easily swapped out for 6mm banana-capable binding posts from Aliexpress (the easiest/cleanest way) or via Clough42's PCB mod that keeps the stock knobs while adding banana jacks.

[Martin72]

Being somewhat of a Siglent fan, I'm surprised you've not mentioned the SDL series.

I´m not a Siglent fan.
Siglent makes good stuff for the money, that's why I have some of it.
But you can't do everything well, I don't know what they're like in terms of supplies and loads.
But yes, the SDL would also be something, especially as it seems quite cheap.
I think I'll buy a ready-made load and rebuild Blackdog's project, I'm currently designing a circuit board for it.
I'll have a look at the loads mentioned, thanks for the input so far.

[s54mtb]

How about KEL2000 series?

https://eleshop.eu/korad-kel2040-electronic-load.html

[nctnico]

How about KEL2000 series?

https://eleshop.eu/korad-kel2040-electronic-load.html

These are quite good (the KEL2010 would fit Martin's requirements) after you move the transformer away from the analog control section. Still, single digit mA accuracy is getting into SMU territory.

Some test data can be found here: https://www.eevblog.com/forum/testgear/korad-kel2010-multicomp-mp710771-review/msg4975759/

[Martin72]

OK,
These three are currently being debated:







Strictly speaking, the smallest Korad is dropped, leaving the second Korad and the Siglent(1020X-E).
According to nctnico's report, I would have to tinker with the Korad to improve it, which somehow doesn't suit me.
I don't know what to expect with the siglent, but I could easily get it and send it back relatively easily if I don't like it.
I would test it like nctnico did with his Korad.

[roy_eedreku]

I had a KEL-103. I liked it for the couple of days that it worked but it just stopped working out of nowhere. The input terminals got shorted and I have no idea how. I was fortunately still in the return window so I didn't tear down or do further investigation. I do have the Siglent SDL 1020X-E right now and it is a great load! I have tested it with my power supply and I was able to accurately pull currents up to 1 mA. I didn't do it for long though so I am not sure about the drift.

[nctnico]

I don't know what to expect with the siglent, but I could easily get it and send it back relatively easily if I don't like it.
I would test it like nctnico did with his Korad.

Judging from a teardown of the Siglent load, it looks like the mains transformer is also relatively close to the analog control circuitry. But in the Siglent load the transformer is rotated by 90 degrees compared to the Korad load which could be better or worse. All in all it would be very interesting to see whether Siglent has a better design to deal with the magnetic field coming from the transformer.

BTW: Tony Albus has created a video showing the Korad, Rigol and Siglent DC loads:

[ArdWar]

I was taken aback by how deceptively hard it is to pick a good DC load.

I once thought that KS EL34243A fits my requirement to a tee. Half rack, dual channel, fast digitization, fast arb waveform, LXI, institutional purchase so no problem with their BS cust support... Yet the only "review" here is basically a trainwreck (for $5K instrument). I wonder if it is still that bad.

[Martin72]

BTW: Tony Albus has created a video showing the Korad, Rigol and Siglent DC loads:

The Korad is "huge", so that could also be a criterion.
From what I can see, and also from your report, it seems to be amazingly well made for the money and the terminals are everything you could wish for.
Korad, you have to remember...
I have now ordered the Siglent load, should arrive later this week.
I'll report my impressions and tests here, I've also "heard" that you can "convert" the 1020X-E to a 1030X, which would be interesting in terms of resolution.
But that would only be an issue if I liked the load, otherwise it would go back.
At the same time, the self-build load will be pursued further.

Edit:

I was taken aback by how deceptively hard it is to pick a good DC load.

Oh yes.
And "good" also means the ease of use, how easy it is to set the desired settings.
That's why the two new Rigol loads are in our cupboard gathering dust, while every colleague uses the ancient Zentro loads because they only have 3 switches and two pots...

[nctnico]

BTW: Tony Albus has created a video showing the Korad, Rigol and Siglent DC loads:

The Korad is "huge", so that could also be a criterion.
From what I can see, and also from your report, it seems to be amazingly well made for the money and the terminals are everything you could wish for.
Korad, you have to remember...
I have now ordered the Siglent load, should arrive later this week.
I'll report my impressions and tests here, I've also "heard" that you can "convert" the 1020X-E to a 1030X, which would be interesting in terms of resolution.

I doubt that is truly possible as the 1030X lists a better accuracy compared to the 1020X-E. Unless the accuracy is crippled on purpose, there must be better parts used in a real 1030X.

I was taken aback by how deceptively hard it is to pick a good DC load.

Oh yes.
And "good" also means the ease of use, how easy it is to set the desired settings.
That's why the two new Rigol loads are in our cupboard gathering dust, while every colleague uses the ancient Zentro loads because they only have 3 switches and two pots...

Skip the Rigols I'd say if that is the case. I have an older TDI DC load (1500W) which has a potmeter to adjust the current but the metering is not very accurate and you never really know where the pot is set to until you turn the load on. At some point I like to replace that one as well but I'm not sure I'd go for a Korad. But then again, it is hard to come up with an alternative that has good performance while being surprisingly cheap.

[Veteran68]

I expect you'll be reasonably happy with the Siglent. It's not perfect, but what is? And yes, it's trivially easy to convert to a 1030X with 3 SCPI commands.

I´m not a Siglent fan.
Siglent makes good stuff for the money, that's why I have some of it.

To be clear, when I said "fan" I meant someone who used and seemed to like their products. I was not suggesting you were an irrational fanboy, LOL.

I myself am a Siglent fan in that I own several of their devices (scope, DMM, power supply, load), but I'm also a fan (so far) of my Rigol DHO1074 despite its rough edges -- at the price I paid, how could I not be? As long as it continues to serve my needs, I'll remain a fan. When it doesn't, I'll get something else. I'm certainly curious to see how the SDS1000HD plays out in the West.

I doubt that is truly possible as the 1030X lists a better accuracy compared to the 1020X-E. Unless the accuracy is crippled on purpose, there must be better parts used in a real 1030X.

Why would you think it not possible? It's been done to death for years now. Of course the accuracy resolution is crippled on purpose, same as with many of these instruments from Siglent, Rigol, and others. There is no hardware difference between them that I know of.

EDIT: You said accuracy, while I meant resolution. Still not aware of any difference between these.

[Martin72]

After comparing the specs it´s only the measure resolution, nothing else.
And more power...Maybe.
And with simple using three commands I could turn the cheapest 1020X-E(522€) into the most expensive 1030X model(1177€).
Sounds too good...Let´s check this before/after "tuning".
This time I ordered from welectron, I know the discount coupon by heart and it was also valid for this load.
So it is still just under 500(incl.VAT).

[electr_peter]

FWIW, Rigol DL3000 electronic load series can also be modified easily (all models have the same HW) to get more power and resolution (color as well), instructions here dl3021-to-dl3031-conversion-discussion

However, I would not recommend Rigol load to OP based on requirements, because old FW with bugs has not been updated for very long time.

[Veteran68]

After comparing the specs it´s only the measure resolution, nothing else.
And more power...Maybe.

Yes, the only difference between the two models is resolution and power. There is no “maybe” as to the power upgrade, it definitely works to upgrade to 300W. I’ve done the upgrade to mine, and there are numerous videos of the upgrade being tested.

[Martin72]

Yes, the only difference between the two models is resolution and power. There is no “maybe” as to the power upgrade, it definitely works to upgrade to 300W. I’ve done the upgrade to mine, and there are numerous videos of the upgrade being tested.

There is still the possibility that the hack only shifted the power limit upwards, or are there pictures of the inside of the 1020/1030 where you can see that it has the same hardware design?

[Veteran68]

There is still the possibility that the hack only shifted the power limit upwards, or are there pictures of the inside of the 1020/1030 where you can see that it has the same hardware design?

I can't point to any photos offhand, but I've seen and read that they're the same. Many have done the upgrade and I've seen reports of tests to 300W.

And there's this post, although not accompanied by any photos, so you'd have to take their word for it (unless there's later photos, I haven't read the entire thread).

I can confirm the 300W model is exactly the same hardware wise as the 200W

I just ordered a 200W and will be converting it to a 300W as my next project

Further I might later look at seeing what safety factor they are using on the main load resistors and upgrade the fan / increase the wattage beyond 300W but that will most likely be a winter project