Seeking Feedback on The Historically Low Number of Applications for Power-Focused Roadtests

Scott,

(hope I didn't get you banned - I clicked Report instead of Reply )

scottiebabe  wrote:

 

Low dollar value roadtests like a power evaluation board could have a streamlined application process for RoadTesters in good standing.

Different opinion: they are a good entry point for newcomers. Maybe a streamlined process for candidates without a proven track?

To All,

Wow! I want to thank everyone for responding and participating in this discussion. It's been helpful. I wanted to go through my takeaways:

1. Knowledge and Equipment: you need both to conduct a power roadtest. We offer test equipment roadtests, but they would not cover all members who need equipment. There are inexpensive university-targeted test equipment, but still I understand that that is not a solution for everyone.

2. Power is not the most popular topic on the community. I have kind of noticed this myself. If members are not writing about power, then it's probably unlikely that they would apply for a roadtest.

3. Some power roadtests are specialized and require more skills.

4. Not a clear personal need for the product being roadtested. For example, multiple rail, higher power products. So, it sounds like I need to be more careful about roadtests on niche products.

5. Less of a need because what boards I use have power supplies on them; I am not designing power supplies.

6. Complex power supplies required specialized knowledge

7. Power isn't the most interesting topic/product out there.

8. The roadtest product needs to have a life after a roadtest. Basically, someone wants to roadtest a product he can use in a project after a roadtest.

9. Product too complicated.

10. shipping delays. Once the pandemic started, we weren't coming into the office. And I think there was some changes inside our office shipping dept. Products clearly were delayed being shipped. We are going into the office regularly now but not full time. We check stock regularly. Also, we have hired dychen  who assists me in all the programs I run, so now I have more time for roadtests. I hope to decrease the shipping delays.

11. Availability and Priorities-- roadtesters have limited availability so they prioritize roadtest applications. Power roadtests are not always at the top of the priority list.

These are all good. Gives me the perspective I needed. If a roadtest does not interest you, I would not expect you to apply. that only makes sense. But some of these thoughts I can improve on. I need to be a bit more careful with niche or complex products. I'm working on selecting people so the kits can be shipped faster.

Thank you.

Randall Scasny

Roadtest Program Manager

I almost didn't apply for the ROHM 8-Channel Multi-Rail DC/DC Converter Board RoadTest, as I didn't want to author a detailed proposal. That being said my application was one sentence, so, it is unlikely I will be selected. So, I question whether I should have even applied at all. My impression is ROHM is looking for a Roadtest demonstrating their voltage supervisor IC. I imagine they would adore a RoadTester if they were to author a Roadtest similar to a recent TI Analog Application Journal: https://www.ti.com/lit/an/slyt814/slyt814.pdf

If I was selected, I only intended to place the DCDC converters in precarious operating modes/conditions and observe the supervisor ICs response. I don't think you need anything you need anything more than a basic multimeter for this roadtest. Now that is just my opinion.

Sponsors could do a better job describing what they would like to see demonstrated in a RoadTest.

Low dollar value roadtests like a power evaluation board could have a streamlined application process for RoadTesters in good standing. I don't think it would be unreasonable to ask that a small number of cut-tape samples are included, so one could make use of product in the future as well. Perhaps, an application section for requests of a few incidental components that would aid in the roadtest as well.

in some cases a gearmotor with encoder  :-)

For some road tests like a motor controller it makes a huge difference to participation if  a motor is included.

Similarly a power module road test may benefit from including something for the module to supply power for.

Hi Nico - Totally agree with you for hobby use,  I tend to use alot of wall warts , more or less switching power supplies  120V to 5v, 9v, 12V dc for protoyping.

I try to stick with 12V so I can then switch the projects to a moving platform like a robot or car, then use lead acid batteries and or 11.1 V LiPo batteries in projects.

I keep at least a dozen LM2596 DC DC 3 amp Buck supplies on hand, a couple XL4015 5A Buck supplies, and a couple 3A Buck Boost for unplanned projects.

I even have a couple buck power supplies with adjustable volts and current limit that I use to put LED floodlights on projects. I have LED outdoor floodlights with failed motion sensors, the replacement motion sensors cost almost the same as a new unit, so the LED floodlights get repurposed.   I set output voltage just above whats needed for turn on volts and current limit as the current adjust.

Actually I've gotten busier, partially because lockdowns, complying with mandates, and remote strategies takes alot more effort to coordinate needed activities, and time to hunt for items that used to be easy to find.

More time spent cooking at home now too.

I'm not saying ripple and noise measurements can't be done with a basic oscilloscope - but to do them properly is tricky and problematic (unless the converter is really noisy).

For example, the issue of the front-end noise of the oscilloscope will be an issue at times, as some scopes are so noisy and have so few bits that you're really reading the front-end noise rather than supply ripple. Using a 10:1 probe gives you the bandwidth but gives a poor resolution of the small ripple voltages due to the attenuation, while going 1:1 results in a bandwidth limitation that depends on the particular probe set (could be ~8MHz) which could lead to values that look too good, and for higher voltage supplies, could mean that your oscilloscope front-end is saturated.

Similarly, depending on the output voltage, you really would like to push the mean voltage down to zero. Many people will whack on the AC coupling option in a jiffy, but this is not actually proper as it will also take out low-frequency DC ripple that may be present. Instead, you should use the front-end offset of the oscilloscope if you can to move it down to the zero line to obtain the maximum clarity of the signal, but unfortunately many oscilloscopes don't have that much offset voltage range to play with.

Because of this, and the fact that many new converters can achieve ripple below 15mV peak-to-peak and 3mV RMS, it's extremely difficult to get a meaningful measurement without a power rail probe. Those are active probes with inbuilt offset capabilities and will DC-couple for proper viewing of even low-frequency ripple. But they are upwards of $2500 and often require a good coupling into the circuit (e.g. via an SMA pigtail). I've always wanted to get my hands on them as I can see my ambient environmental noise and more affecting my measurements on the regular 10:1 probes.

- Gough

Gough Lui  wrote:

Evaluation requires experience and equipment. For power conversion, some of the more interesting measurements such as ripple and noise and EMC require highly-specialised equipment and expensive probes.

EMC certainly does require specialised equipment, I've done ripple/noise measurements with just a basic scope.

What is tricky is finding the PWM frequency from an unknown PSU.

Time yes, patience no (not being able to go out much/at all, does that to people).

Sit around for a month for a RoadTest to close, wait a bit longer for them to decide, find out they selected someone else, slowly go crazy (optional).

Plus your forgetting the many issues with shipping products around the world, so long wait for things to arrive.