Question about AIM-TTi LD400P DC Electronic Load inrush current

[eyiz]

Hi guys,
I'm about to pull the trigger on one of my quotes and buy the AIM-TTi LD400P DC Electronic Load for my home lab.

However, something someone mentioned to me while researching DC Electronic Loads is that some of these types
of equipment are only suitable for industrial use, and won't work in a residental circuit that is fused for
15-amps at 110-120V, which is my situation here in Canada.

For example, I looked at the new KIKUSUI PLZ405W, which is above my current budget, but seems a really nice
instrument too, that has some of the specs similar to what I was looking for, and found it's specs state that
the inrush current is 45-Amps, even though the unit is only 50VA on the mains input, so typically pulls less
than a 50 Watt light bulb from the mains, and Kikusui specifically states that their instrument is suitable
only for industrial use, probably partly because of this, and also because of possible EMC
issues stated in the footnotes of their manual on pg.108 here;

[pic]

Now, this all got me wondering about the AIM-TTi LD400P, whether it also has a high inrush current. I couldn't
find any information in the TTi manual or datasheet to clarify this issue. It seems the AIM-TTi equipment is
not as well documented as the KIKUSUI equipment. I contacted AIM-TTi through their
webform and enquired about the inrush, and asked whether this unit is suitable for a 15-amp fused 110V-120V
residential circuit. But, while waiting for a response from them, I thought some more knowledeable folks around
here could help answer the question.

Does anyone know anything about this? Anyone familiar with the LD400P? Can you use the LD400P on a residential
15-Amp circuit, or will it blow the mains fuse, or circuit breaker, everytime you start up the unit, because
of the inrush spike in Amps?

[bugi]

While not an expert on these issues (or on anything for that matter), I can say that I have 400W and 500W PSUs with stated inrush currents at 70A and 40A respectively (with "active inrush limiting", pfft), half of those values on 115V, and tested results indicate getting near there. And they all work fine on a 230V 10A circuit breaker. (I have not tested to turn more than one on simultaneously, yet, though I have turned on one 70A unit simultaneously with my oscilloscope.)

The thing is that the inrush lasts very short time, so slow fuses or circuit breakers can usually handle the total inrush energy. I calculated the values for the other model, based on the input elements they have, and estimated that with the stated peak current, the inrush should last less than half a cycle and the peak of it even much less that.

However, note that this was the case for those particular PSUs; it all depends on the total energy needed for each device, and particular household fuse/CB etc.. Hmm.. the LD400's spec states 6A minimum mains lead.. I would assume that if they have specifically notified about that rating, they would also mention about fuse, if there was some significant requirements for it.  But assumption is the mother all ********, so...

[Fgrir]

I have an LD400P in my office, which is actually one of the bedrooms in my house.  The power is coming in through a 15A AFCI breaker which also powers another bedroom, but I've never seen even a hint of a problem when turning on the LD400.

[MarkL]

I had an LD400P for a couple of weeks and used it on a 15A circuit.  There were no issues at all with in-rush.

I don't have it anymore, however, because the loop response was poor (read: unstable) compared to other loads in identical scenarios, and the LCD user interface was so badly designed it made me curse out loud at it.  I have not used any Kikusui loads, so I can't say how it compares.

If you decide to get an LD400P, I'd recommend you get one from a distributor that will accept a return if the unit doesn't meet your application needs after an evaluation period.

[eyiz]

I had an LD400P for a couple of weeks and used it on a 15A circuit.  There were no issues at all with in-rush.

I don't have it anymore, however, because the loop response was poor (read: unstable)

Thanks for that heads-up MarkL.

One of the claims made by AIM-TTi is that this LD400 Series is a true linear slew. So, the ramp up from one level to another is not a series of small steps, like some other products. Did you examine this feature, and can you confirm their claim that the level changing is smooth and linear function of time?

[eyiz]

I didn't think myself that inrush should be a problem with LD400P. But, I just don't know it for a fact.
The way I analyzed it was to look at the specs for the TTi to see how fast it can change levels.

As a general rule of thumb, the faster the change the greater the overshoot and the higher the spike produced, and also the greater the intensity of the burst of electromagnetic radiation emitted, even if for a short time. That's because, at the most fundamental level, Maxwell's equations tell us that the magnetic field created is proportional to the time-rate-of change of the electric field, and visa versa,
curl B = dE/dt, curl E = - dB/dt, etc..

So, anytime you change electrical parameters "more quickly", you've generally got to deal with higher spikes and more
intense bursts of EM.

The KIKUSUI is in a class of it's own, boasting a whopping 50A/us for PLZ-4W and 60A/us on newer PLZ-5W series. Whenever you have circuits designed for that kind of responsiveness, you've got to expect spikes when switching on these circuits, or even switching levels while on, in normal operating mode. The whole system must be designed to exhibit this fast responsiveness throughout the circuitry. That's because, in a relay race, the slowest runner determines the limit of overall speed of the team. I seriously doubt, that the KIKUSUI can do all this in a purely linear manner like the AIM-TTi. But, then again, I don't know.

Now, to compare, I looked at what I could find in the LD400 documents to get some idea of TTi's responsiveness.

Here's some relevant specs from pg.4 of the LD400 Series manual:

Specifications


If we convert the A/s in the CC mode shown to A/us to put the numbers on the same basis
as the KIKUSUI quoted 50-60A/us values, we get:

8A range: 2.5A/s to 250A/ms  =or= .0000025A/us to 0.250A/us
80A range: 25A/s to 2500A/ms =or= .000025A/us to 2.500A/us

Highest rate of change for AIM-TTi LD400 Series is 2.5A/us
Highest rate of change for Kikusui PLZ-4W Series is 50A/us
Highest rate of change for Kikusui PLZ-5W Series is 60A/us

So Kikusui is more than 20 times faster than AIM-TTi at the LD400P's fastest speed, and more than 200 times faster than LD400P on its 8A range.

The AIM-TTi LD400P is like a Tortoise on the race track, while Kikusui is the Hare.

But, there are always design tradeoffs to make when attempting to construct any kind of circuit. I'd suspect that AIM-TTi LD400P has no spikes, overshoots, etc.. to be worried about, so would be relatively safe to hook up to almost any circuit,  that might have sensitive components that could be affected by such things, but  wouldn't be able to keep up with some of the modern fast changing circuits out there.

While, the Kikusui might happily handle anything you throw at it, but probably wouldn't be nice to some kinds of sensitive circuit components, especially if you didn't know exactly what you were doing on each occasion. e.g. a hobbiest like me.

So, a back of the envelope quick guess, or Dave's thumb, would suggest that  perhaps the LD400P should have 20 times smaller inrush peak than the KIKUSUI, or 45A/20 = 2.5A peak.

Of course, there's nothing that says the inrush is linearly related like this, but, lacking further knowledge about the design, it's a best efforts guesstimate. I'd hardly think that the inrush would be higher than the Kikusui, but who knows? It could all come down to the efficiency of the design in achieving the design goals.

But, this slow response of the LD400P vs Kikusui also supports MarkL's observation that the LD400P can't keep up with the time-changing parameters for loop response tests. The LD400P is a different type of beast, intended for low voltage and slow changing parameters, like solar cell testing, for example.

I notice that Bob Hanrahan of Texas Instruments used a Kikusui PLZ164WL in his power supply testing instructional video series. That's where I first became aware of the Kikusui. If texas instruments chose this instrument for their testing, it must meet or exceed the specifications required for such design testing. However, I can't afford such luxary equipment.

I considered the AIM-TTi LD400P, because I bought the AIM-TTi MX100TP and really like this three-channel power supply. So, I figured, let me stick with AIM-TTi and get the DC load too. It's in the price range of my limited budget.

MarkL, you mentioned that other DC Electronic Loads in the same class as LD400P were better at loop response tests. Could you elaborate a bit, what other brands and units did you comapare this AIM-TTi produt to? Are they in the same price range?

 

 

[MarkL]

Like you, I also have a Aim-TTi MX100TP power supply and I also have their other model, the MX180TP.  Both have been excellent performers.

I already had an Array 3723 load, but thought I would try the LD400P given the track record of their power supplies.  I needed a little more load power for a project than the Array could handle.  I didn't have a list of other loads to evaluate, nor did I have a lot of time.

After it was clear the LD400P would not do the job, the testing phase of the project was modified to allow two loads to be used, and in the end I purchased another Array 3723.  It's in the same price range.


I thought I had saved some of the scope screenshots of the LD400P's behavior, but I unfortunately can't find them.  As you've discovered on their spec sheet, it was exceedingly slow.  I was using it in constant resistance mode and that's where I had the most trouble with significant under- and over-shoots and at times oscillation.  (Besides being driven crazy by the user interface; it is NOT the same as their power supplies despite the similar looking front panel.)

I also noticed that in CR or Constant Conductance mode, it only had two slew rates it actually used.  It chose which one it was going to use based on Slow Start and some other settings (sorry I can't remember the specifics).

But still, the LD400P may work ok for you in Constant Current mode, which seems to be your focus.  When operated slow enough in CC mode, I never noticed any stepping from level to level.

The only other loads I've evaluated, and this is a couple years in the past, are BK loads.  I will never touch another piece of BK equipment because of their terrible technical support.  But they continue to be popular and are also in your target price range.


Unless you're doing something really simple, like battery discharge curves, I think there's too many variables in the way these loads are designed to know how they'll really behave in a dynamic situation without an actual evaluation.

[bugi]

About the inrush.. In the first post you talked mains and fuses. The "load" side of these devices (well, the device itself is a load, but the side that is being controlled, 80A, 80V, slew rates etc.) should not have much anything to do with the mains input side, especially not with the inrush current. The mains input should mostly be used to run the control circuitry, display and such.

Thus, you can not deduce anything from the load control specifications the way you tried to do. Well, indeed, as you estimated, fast and high changes in current will do noise, and that noise could couple backwards into mains at some level if the unit wasn't designed properly. (But still likely less than the noise already on the mains due to all the other devices with their cheap noisy PSUs screaming back to the mains, unless you have added extra filters on them all, or have them on fully separated mains supply.)

The inrush current depends much more about what kind of transformer its PSU uses, or if its a switcher, then how big capacitor(s) it has, and even more on what kind of inrush current limiter it might have (if any) and its parameters.It could be a simple but relatively big resistor with a parallel relay, or just a single NTC, which is always a compromise between inrush limiting and steady-state inefficiency, etc. etc.  One design might use 200ohm NTC, another 600ohm, and that alone would make about 3x difference for the peak inrush current (but note, not for the total inrush energy).  (PFC feature could also handle inrush limiting, sometimes in awesome levels, but looks like the device doesn't have it.)

Note: http://resources.aimtti.com/manuals/LD400+LD400P_Instruction_Manual-Iss2.pdf  page 11 end:

... an internal fuse on the
power supply PCB, which is intended to protect the unit from the accidental connection of 230V
mains supply to a unit configured for 115V operation. Before replacing this fuse, ensure that the
unit is configured correctly, as described above.
The replacement fuse must be a 20x5mm 500mA (T) 250Vac rated HBC (ceramic tube) type.

From this we could estimate that at 230V, a 1A slow (T) fuse should be able to handle the inrush current. Way less than what normal household mains fuses are rated at.

I suggest reading that manual if not already done so, it seems to contain all kinds of bits of info about stability, transients, etc.