Re: Using cordless tool batteries to power a drone

I've been shopping for a clamp type DC ammeter to replace various lesser quality devices which I use for measuring DC stick arc welder current, spot welder DC peak current, automobile charging current, DC starter current, LED flashlight current, radio transmitter DC current, solar panel short circuit current, etc. I didn't know that the MS2018a would measure DC current through the clamp. Looks like it should work. However, I'm tempted to buy a Mastech MS2109a, which is similar but has a better front panel layout and adds a thermocouple thermometer and NCV (Non-Contact Voltage) AC tester for approximately the same price.

I haven't found a way to filter Google or eBay searches for clamp meters that will work with DC through the clamp. Most meters will do both AC and DC current through the test leads, some will do AC through the clamp, but very few will do AC and DC through the clamp. One trick I've found is to look at the silk screening on the ammeter ranges. If it shows a stylized sine wave over a solid line, it should work with both AC and DC. The clamp also has + and - markings.

Very little voltage drop. A 10A 100mv shunt has a resistance of 0.01 ohms and will drop, umm... 100mv at 10amps:

I have some opinions on what's wrong with the quadcopter. However, I'll wait until I see some voltage and current measurements and graphs performed in a test stand (like your chair test). My guess(tm) is that this sounds like a high resistance connection between the battery and the ESC (electronic speed control). Something seems to be prematurely announcing that the battery is depleted. An unwanted voltage drop in the wiring would do that.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

I've got two Uni-T meters

both 4000 count models, one a true RMS autoranging DVM and the other a clamp ammeter DVM, model UT204 which does AC and DC and autoranges between 40A and 400A scales. I've used it to check starter draw on a car and the output of a 10W solar panel (0.22 A that day) and I'm happy with it. They have lots of models with different features, and lots of sellers on amazon and ebay. The Uni-T meters seems to have good accuracy and useful features, but get their low cost from slow settling times. Over a second per screen update and often three updates to get a stable reading so 3-5 seconds total. For my needs that works.

If you don't use a clamp-on meter be sure to check the manual for your meter. Lots of the cheaper meters with a 10A scale bury a truly wimpy duty cycle in the fine print. I've seen 15 seconds of 10A measurement followed by ten-plus minutes of required cooldown in at least one of the Uni-T meters and some other brand I can't remember right now.

--
Regards, 
Carl Ijames

MS2109a data sheet:

More Mastech clamp meters:

True RMS would be nice to have, but is not necessary for what I expect to be mostly DC current measurements.

Sigh. I went to the Uni-T web pile and tried to look for meters. Every link link I click returns a "404" (nobody home). Despite my having an uncontrollable urge to buy something immediately, I'll wait a day or two for someone to fix the Uni-T web pile or server.

Thanks. I'll do some reading and see what they have to offer.

Argh. I didn't realize that this was a problem with DC clamp meters. I just skimmed the Mastech MS2109a manual but didn't find any mention of a cool down period. For sampling rate, it claims: "Sampling rate: about 3 times/s, simulation bar 30 times/s." Incidentally, the highest count is 5999 which is needed for the 600A scale.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

I can see that I'm not getting through to you. Please excuse my abrasive, rude, undiplomatic, and short tempered manners.

Here's what you should be doing:

1. Setup your drone in a manner similar to Jan Panteltje's test chair:

1. Take DVM and place it across the battery terminals as close to the actual battery as possible. You want to measure the actual battery voltage.

2. Take a 2nd DVM and place it as close to the place where the ESC distributes the power to the various motors. That would be the DC input to the ESC. If the low voltage measurement device is located elsewhere, use that connection.

1. If you have an adapter for your XT60 battery connector for current measurement, attach it and also record the current draw ever 30 seconds.

2. Run the drone in the chair at something resembling normal flight RPM (not full power). With a paper pad and pen, record the voltages every 30 seconds. When the low voltage alarm is triggered, record the time. Let it run until the controller or ESC declares the battery to be discharged. Draw a graph. If my guess is correct, the graph will look very strange and that there will be a substantial voltage difference between the battery terminal voltage and the ESC voltage.

Really. Have you calculated what your runtime should be based on the average current drain and battery ma-hr rating? Be sure to check the data sheet for the LiIon cells for the ma-hr capacity at whatever current your motors are drawing. How close is your flight time to the theoretical runtime?

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Jeff Liebermann wrote

Yes, those things are getting better all the time, I notice that if I google 'MS2109a user manual' then I get all the info on that thing as pdf.

Indeed, external shunt, but if he uses a multimeter meter those test leads are very thin and crap, for current, you will need short thick test leads to the meter,

For the original lipos the wiring is very short, big connectors, don't know his setup.

Jeff Liebermann wrote

I can confirm that for the Mastech MS2108, I can set the current limit for my battery charger with it and it reacts as fast as I can turn that 10 turns trimpot.

Do not know about a cool down period, never measure high DC currents for a very long time. Before measuring I press the zero button once. And I did put 3 eneloop 1.5 V AAA batteries in it so it is always ready.

Don't understand this, I checked their site just before posting my first reply and again just before posting this, and had no problems.

I had my other meter in series and both agreed at 0.22A within about 0.01 or

0.02A. I was pretty impressed how low the clamp on could go, 40A and 4000 counts makes the LSD 0.01A.

I guess I wasn't clear enough, the non-clamp meters can have the duty cycle issue but so far as I know none of the clamp on meters have a problem - that's whey I said "If you don't use a clamp-on meter ...".

--
Regards, 
Carl Ijames

What browser and OS are you using?

I just checked with Firefox 52.9.0 on XP (same as Jeff) and Firefox 64.0 on Win7.

Both gave 404 errors on all links.

Can't even contact them to tell them their site is down.

How are you able to get pages?

Firefox 64.0 32 bit on Win 7 Ultimate 32 bit. I may have updated Firefox between my two posts, or maybe after the second one but definitely 64.0 now and it works (can't remember exactly when I told Firefox to update vs. posting), and the previous version was the most recent one before 64.0. Weirdly, I clicked on your contact link and got a 404 page in Chinese. Then tried just "

" and got the home page in english, as always.

--
Regards, 
Carl Ijames

Thanks for the reply.

I just tried Firefox 64.0 on Win7 Pro 32 bit. Same result using "www.uni- trend.com". 404

Wierd.

Yep, that MIGHT be a problem. Let's do the numbers.

Discharge curve for a AAA Eneloop NiMH cell: compared with a Duracell CopperTop alkaline AAA cell:

According to the MS2108a manual (Pg 18), the low battery indicator will complain at 3.7V or 1.23V per cell. I don't know how much current the MS2108a will draw from the battery, so I just measured the current drain on a Radio Shack 22-178 that runs off two AA cells. I measured 330 microamps, so a 3 cell arrangement should draw 220 microamps.

Inspecting the graphs for the two types of batteries, and extrapolating the family of curves to something around 220 microamps, I find that the alkaline will last >900 hrs before the voltage drops to 1.23v, while the Eneloop will last about the same >900 hrs. So, no benefit in operating time.

Running the backlighting in the MS2018a might increase the drain to maybe 10ma, which might cause the battery low warning indicator to appear. I suspect that will only a be a problem with an almost depleted battery.

However, the alkaline cells leak (badly) and are not rechargeable. Eneloop cells are LSD (low self discharge) and will be ready when you need to make measurements. I've been systematically replacing my various AA and AAA cells with Eneloop NiMH cells, and my 9V batteries with rechargeable LiIon equivalents.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Probably the Great Firewall of China at its best. I have 3 machines in front of me. XP, Win 7, and ChromeOS. XP and Win 7 have Firefox, Chrome, Internet Exploder, and Vivaldi running. XP and Win 7 show: 404 - File or directory not found. The resource you are looking for may have been deleted, the name has been changed, or it is temporarily unavailable. on all browsers as translated from the original Chinese by Google Translate.

I also cleared the Firefox and Chrome browser cache and reloaded with . The Acer Chromebook 14 CB3-431-???? only runs Chrome, but it too returns a 404. I thought it might be an ISP issue, so I connected via the neighbors Comcast router. Same 404.

I also tried but it would not connect. I also removed all cookies associated with uni-trend.com, but the 404 error persisted. I also clicked on most everything possible on the home page. Everything I tried resulted in a 404 error.

Next, I did a fast port scan using nmap (actually Zenmap). The server is running Microsoft Windoze Server 2008 Beta 3. Beta? Nothing like 10 year old software. My best guess(tm) is that this problem lives in the 10 year old server software:

# nmap -T4 -F

Starting Nmap 6.46 ( ) at 2018-12-16 14:01 Pacific Standard Time Nmap scan report for (59.188.250.13) Host is up (0.31s latency). Not shown: 97 filtered ports PORT STATE SERVICE

80/tcp open http 135/tcp open msrpc 49154/tcp open unknown Nmap done: 1 IP address (1 host up) scanned in 7.64 seconds

Looks like they have a web server with WMI (windoze manglement interface) enabled to the internet. I can go on with this, but I don't feel so good today and would prefer to do something else.

Oh, now I see. I was wondering what might be getting warm when using a clamp type ammeter. Having smoked a few DVM's in the distant past, I don't like running high current through them. I have a small collection of shunt resistors that work quite well.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Just to clarify things, I was able to get the home page at: on all my machines and with all browsers. The home page worked. What doesn't work for me are any of the links and buttons on the home page.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

I thought you meant that the home page gave you 404. Before my first reply I followed several links from the home page to get specs on a few meters, and all worked as expected. Now, after reading your post I tried that again and now I get 404 everywhere, too. I even went back into my history and tried to go to a couple of the pages I visited yesterday and now they also give 404. So, now I can't get anything at their site, either. I have no idea how to make progress so I think I'll give up until the next time I need a meter :-). Good luck if you decide to keep trying.

--
Regards, 
Carl Ijames

Jeff Liebermann

is working OK here in Europe, automatically changes from Chinese to English somehow. Linux Seamonkey, but the menus give a Chinese text and error 404. It is not your browser

BTW before I bought that MS2108 I did some googling and watched some youtube videos about both the uni-trend and the Mastech meters, the choice was not hard to make. It is funny if you see somebody advertising a meter by comparing it to an other one and the scale tips the other way all the time :-) Hey, I have no shares in either ..

On Dec 16, 2018, Jeff Liebermann wrote (in article):

For me (Boston area), works but does not. The difference is http versus httpS. If the Chinese want to listen in, it makes sense to forbid https.

Joe Gwinn

John Doe wrote in news:q6ib0b$81p$ snipped-for-privacy@dont-email.me:

I would make a front end circuit that can still make output even as the input sags down.

It seems you drive circuit senses too low a voltage and stops asking, and then the batteries never go below a certain point in their discharge curve. A switcher between the battery and the circuit input can fully deplete the batteries while providing the voltage the circuit demands for a longer period.