Hi All,

Here's my story;
I have been asked to manage a project for a company that claims to have the technology to produce the most power efficient intelligent chargers and converters on the market today.
The advantages are huge: Less energy used ($$$), less heat, longer life, tighter packaging, reduced cooling requirements, etc...
The downside is 20-50% higher cost, which is why this company has not been able to move their technology for the past 15 years. No one cared about efficiency until now and the only determining factor has been cost.

Clearly, efficiency is a hot commodity these days and I am interested, but before I can begin investing time and resources into this project, I wanted to verify their claims, so they have sent me a prototype charger. Now, it sits in my house and I am looking for a company that can test the device and produce an independent report on its performance.

I would greatly appreciate any help and advise!

Just for your info, the claims are as follows:
Test parameters
The following parameters were obtained during testing between 500W and 1000W:
1. Output voltage: +410 VDC
2. Output Power: 0 - 1000W
3. Input Current: 0-265VAC, 45-400*Hz, 1 phase, 100-340VDC
4. Efficiency: 96% at 110VAC, 98% at 220VAC
5. Power Factor: 0.99* at 110VAC, 0.98* at 220VAC
* Figures obtained during testing at 50Hz.

 

If I may ask - what do you do with 410 VDC @ 2.44 amps? Forgive me, but that sounds like a trickle charger for a diesel submarine.

Something lower in voltage would seem to have more potential applications.

 

If I may ask - what do you do with 410 VDC @ 2.44 amps? Forgive me, but that sounds like a trickle charger for a diesel submarine.

Something lower in voltage would seem to have more potential applications.

Good question.
The technology has been originally developed for commercial applications. Its efficiency is proportional to the power output. However it can be scaled down as well. The closest application that I envision is charging of electric vehicles.

 

The voltage may be right for an electric car, but the charging current is an order of magnitude too low. Will the power scale up?

 

The voltage may be right for an electric car, but the charging current is an order of magnitude too low. Will the power scale up?

Yes.
I am not aware of any limitations up or down.

 

I'd recommend you find someone with a DC load instrument -- they'll likely be able to make the measurements you're after. It sounds like you're in the UK; I wouldn't know who to recommend there.

If you were in the western US, I might be able to help, as I've done similar evaluations on smaller DC supplies. My DC load only goes up to 300 W, but I might be able to get my hands on a larger one.

The instrumentation needed is straightforward: a scope and shunt to measure the input AC power (I assume it's single phase at these low power levels) and a DC load to apply a load to the DC output. A computer is handy if you want a load that changes its characteristics over time. If you could rent the load and have the other equipment, you could do the measurements yourself (they're not terribly difficult).

However, if you want to evaluate the device at multiple power line frequencies, that gets more complicated and you're likely looking at taking it to a testing lab and paying a goodly chunk of money.

I'm intrigued by the efficiency claims. If someone says to me something is 98% efficient at 1 kW output, that means I'd measure 1.02 kW real AC power input for 1 kW output. This means only 20 W in heat lost internally in the device. I'm no EE, but that sounds pretty challenging to do if there's a conventional transformer and solid state switching devices inside. Or are their marketroids' fingerprints on the numbers?

 

I'd recommend you find someone with a DC load instrument -- they'll likely be able to make the measurements you're after. It sounds like you're in the UK; I wouldn't know who to recommend there.

If you were in the western US, I might be able to help, as I've done similar evaluations on smaller DC supplies. My DC load only goes up to 300 W, but I might be able to get my hands on a larger one.

The instrumentation needed is straightforward: a scope and shunt to measure the input AC power (I assume it's single phase at these low power levels) and a DC load to apply a load to the DC output. A computer is handy if you want a load that changes its characteristics over time. If you could rent the load and have the other equipment, you could do the measurements yourself (they're not terribly difficult).

However, if you want to evaluate the device at multiple power line frequencies, that gets more complicated and you're likely looking at taking it to a testing lab and paying a goodly chunk of money.

I'm intrigued by the efficiency claims. If someone says to me something is 98% efficient at 1 kW output, that means I'd measure 1.02 kW real AC power input for 1 kW output. This means only 20 W in heat lost internally in the device. I'm no EE, but that sounds pretty challenging to do if there's a conventional transformer and solid state switching devices inside. Or are their marketroids' fingerprints on the numbers?

Thanks for your reply!

I am actually in the US. New York, to be more specific.
I tried searching for a lab that would do independent testing, but came up blank. There are labs that test all kinds of things, but several that I've contacted told me that they don't do this kind of testing

As to the claims, the person who has developed this technology has patented it and it appears to be legit. The theory behind it is waaay over my head, but from what I understand, and this is a gross over-simplification, it combines multi-stage conversion process with PFC to come up with such a small loss.
I have the first prototype that has been put together on a budget, meaning that not the best components have been used and there is room for improvement. However, engineers who introduced it to me have assured me that a knowledgeable EE will be able to see the advantages regardless.

I am willing to pay money (within reason, as it will come from my not-so-deep pocket) for a lab or a university to conduct a test and produce an independent report. If you can suggest whom to turn to, I'd greatly appreciate it!

I am also looking for energetic people with connections in the EE industry to help me market this technology. If the claims can be substantiated, this can be huge, given todays realities.

 

You could contact one of the technical schools in a large industrial town/city, they usually have good DC machines labs full of high power test gear, for training electrical technicians to work in mining and heavy industry etc. A 1kW DC charger is not big for them.

And please post your results if it's not too embarassing, I really can't see it being 98% as anything over 95% is really hard once you get large currents involved... Every connector/PCB trace/inductor/busbar/FET has a couple of milliohms and the I2R gets you every time.

97% at 100W is do-able but costs a lot $$ in very special construction, over say a more typical 92% just using good design and quality parts.

But 98% at 1000W?? As I said, please post independent results.

From a business perspecive you are probably better off making a 92% efficient product for $100 manufacturing cost, instead of a 97% efficient product that costs $500 to make. The customer won't care that much, they will most likely buy the unit that is reliable and has the lower retail price.

 

Thanks for your reply.
I will definitely post results once I find someone who can test it for a decent price. So far, I've been quoted $3700 for the test and I think that it is way too much for what I need done.
I don't need the kind of comprehensive test that a product must undergo prior to being sold to the public - Just a proof of concept to verify the claims.
Noone seems to be offering this for some reason.
I keep looking.

 

You say they have patented technology -- get the patent numbers and post them. It only takes a few minutes to read through a patent to estimate if they have any useful claims. From experience, be wary if they refuse to give you the patent info or dance around -- they're almost always charlatans.

I'd be happy to make the measurements for you (inexpensively, as I'm a retired HP engineer/scientist), but I don't think it would be economical to ship the thing across country (I live in Idaho). You could make the load measurements yourself if you have some measurement experience and an oscilloscope. Renting a DC load would be the least amount of work; there are also ways to build a load yourself. I know a person in Bayside, NY who might be willing to make the measurements for you with my help. PM me if you want to talk further on this subject.

Even if the technical claims are true, the real challenge here will be the marketing and sales. I would insist on seeing a carefully-done business plan that focused on competitive advantages, competitors, and who's going to buy this and why they'd choose it over other stuff. If you're the one who has to develop the business plan, then you've got a lot of work ahead of you (i.e., you need to do your due diligence). Personally, I would focus on customer benefits first. You'll have to show why a tight-fisted businessman will spring the extra bucks for this technology. The only reason will be that it saves them energy costs AND gives a pay back in a reasonable amount of time over the competition's equipment. Thus, you have to be prepared with a graph of ROI time versus electrical energy cost -- and be prepared to compare it to existing technology that might be 5% less efficient. You'd definitely get a yawn out here in the northwest where electrical energy is pretty cheap, but back east or in California I think people pay 3 to 5 or more times as much as we do for electrical power, so there might be a market. It will hinge on payback time.

A knowledgeable engineer would also be leery of some new technology because it may be untested. You'll have to show him hard data on cost of ownership that shows he will benefit and not be looking at unreliability problems down the road. An experienced engineer or manager knows that unreliable equipment winds up costing many times the original purchase price in lost revenue, pissed-off customers, etc. This may be the biggest sales barrier unless the monetary savings are large -- which they won't be, not for a 5% improvement in efficiency.

If you can't show him that hard data, then you're asking him to take a risk -- and the only thing that might lower that barrier is to lower price and/or guarantee minimum downtimes (and you have to get him to believe you). Making a mistake here can cost you the business (i.e., put you out of business).

 

Thanks for your reply.
So far, I've been quoted $3700 for the test and I think that it is way too much for what I need done.

I agree; that's way outta line. Here's an outline of what I'd do:

Hook up a scope to measure the input AC current and voltage. Hook up a DC voltmeter to measure output voltage and use a clamp-on DC ammeter to measure the output current. Once the thing was turned on and had reached apparent thermal stability, I'd take a measurement. I'd test at both at 110 VAC and 220 VAC.

The only uncertainty is that I'd build my own load. I have two designs for this. One is easy to build with parts on hand, but how well it would work is uncertain. The other design is almost certain to work well, but involves ordering parts and some work in my shop to build. I would charge the same time and materials for the second design as the first (even though it would take substantially longer and cost more for materials), as it would be something I could use for other projects. The first design would take perhaps a half hour of shop time and a half hour of testing.

Frankly, it would make more sense to find a retired engineer/scientist near you that has the equipment and have him make the measurements. Then you wouldn't be looking at freight charges. But if you can't find someone locally, send me a PM and we can discuss things in more detail.

 

To "verify the claims" you are probably better off seeing a university or technical college. They will have standard procedures, standard calibrated shunts and voltmeters etc all of which can be written up in the procedure showing which tools they used in the testing in a lab type environment, so the tests could be repeated by others if they challenge the claims.

Even though a backyard engineer is probably as smart and may be just as well equipped with the tools, it still won't have the marketing cred of a uni or college test. There should be a teacher/lab worker etc out there that can do the tests in his lunch hour and write it up for you for a few hundred dollars.

 

You only need a big resistor or a small one immersed in running water.

The scope is needed only to observe the output to confirm it does not distort badly under load.

What is definite and unarguable is the two wattmeter method. One for input AC power and one for output DC power, with the DUT (device under test) in between.

One don't need to know what DUT is or how it works. The claim at the rated output power is verified if output power is the claimed percentage of input power, as calculated by the readings shown on the wattmeters.

Every college or technical institute should have these wattmeters and I think most will be happy to let you do a demonstration to their students as a practical test to verify an industrial product.