EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: sairfan1 on January 12, 2021, 08:36:33 pm
-
I'm working on a ESP project where I want to use my ESP8266/32 with different batteries, to calculate accurate current consumption and battery life i was doing some research for a suitable tools, what i really want a graphical interface (via PC software) where i can see variation in current consumption from nA to uA and mA along with time span.
I came across uCurrent Gold as its about 11 years old project not that user friendly and looks like Dave is in process of improving it, I'm not considering it for now. (any idea when it will be available?)
Current Ranger looks good but looks like it has some limits while sudden increase in current? max current range is around 300mA and its PC interface also has some issues, anyone here can explain their experience?
I watched Dave's video regarding BattLab-One (BlueBird Lab) this tool has great price, but it does not go that down to nano amps and needs lots of improvements for now.
I also came across some expensive ones like OTII and Joulescope not sure how important it is to buy such expensive tool, is it significantly important to have such tool for easy and accurate calculation?
any other tools from ebay that can be used like those power supply or meters like HP3457A, I can spend around $200 but if its significantly important to have tools like Otti/Joulescope then i can think of that.
-
Please state ESP32 or ESP8266. Just mentioning ESP makes me think of ghostbusters. :D
Said that, depending on the frequency of the change in magnitude of the current (duty time) and desired accuracy, I doubt a general DMM like the HP3457A (which otherwise has attractive features) will be of much help. Maybe more recent DMMs which are much faster could, but they tend to be quite a bit more expensive.
I'd think for battery lifetime investigations, some kind of integrating scheme is more appropriate. After all, one really cares about energy, not so much power in that context.
-
I will be working on both ESP8266 and ESP32
I have HP3457A sitting in my lab but did not work on it, sorry it was just a hobby purchase :) and i was thinking it could be helpful, does it has some software support? to see change in current graphically like uA/time
Desired accuracy is something like if i calculate 365 days it should not die in 340 days but 5 days +/- is ok, based on current consumption i want to further fine tune things to get max time.
it would be great if i could use such investments in future projects like using NRF based sensors, ESP based small motor control etc.
-
I had never had much interest in the uCurrent. It's not much more then some shunt resistors and an opamp in 100x configuration and I'd much rather build something myself than buying that.
Low current applications are usually battery voltage level circuits and it's not so difficult to build a precision power supply with a 1V to 10V output and upto 100mA or so of output current. For such a power supply you can use a relatively high value shunt resistor for current measurement, and then just let the power supply itself compensate for the voltage drop of the shunt resistor.
When you do it this way you also do not have to use an extra box with extra batteries and extra wiring.
Overall it looks like an under represented market segment. There are loads of power supplies in the 30V 3A range with bad quality meters so you have to add DMM's to get some decent resolution, and there are some extremely highly priced precision instruments such as the Keithly SMU (Some EUSD4000 or more).
-
From my experience ESP32 tends to be very sensitive to weak supply rails and will boot-loop if the voltage drop/ripple is too high.
So you want something with low burden voltage.
I had never had much interest in the uCurrent. It's not much more then some shunt resistors and an opamp in 100x configuration and I'd much rather build something myself than buying that.
µCurrent's main selling point is it's low burden voltage and if i remember correctly Dave even mentioned that one of the main use cases for the µCurrent is to measure the current consumption of microcontrollers in sleep mode.
You can make one but Dave's µCurrents are all tested and guaranteed to work and be accurate.
-
does anyone has idea when Dave's new uCurrent is coming in?
I do not want to spend time to build my own current measuring solution, rather i want to focus on device development and looking for some solution for current measurement so that i can fine tune it.
-
does anyone has idea when Dave's new uCurrent is coming in?
I do not want to spend time to build my own current measuring solution, rather i want to focus on device development and looking for some solution for current measurement so that i can fine tune it.
You might want to PM Dave, or go to the Buy/Sell/Wanted section to look for one.
-
What's wrong with using a shunt with your HP? If you want some automation then take the money that you were going to spend on some hardware and get a HPIB to USB instead, which you might find handy down the road for something else anyway.
-
What's wrong with the uCurrent? I use one to measure current flowing through liquids containing nanoparticles. It works like a charm and isn't "difficult". You connect it in your circuit and hook its output to an ADC. I sample it at 32kHz. I don't understand. In fact, having the uCurrent in my instrument revealed something so unexpected and previously unknown that I ended up with a patent. Cheers, Dave and Go uCurrent!
-
@msat
What's wrong with using a shunt with your HP?
I'm not an experienced developer, did you mean? what i understand, I can shunt a resistor to my HP3457A to read nano range? can i get those reading directly to my computer to log?
you were going to spend on some hardware and get a HPIB to USB instead
How this hardware can help me?
@JohnnyMalaria
What's wrong with the uCurrent?
What I understand, I have to switch modes between nA, uA and mA that is not possible in real time scenarios, for example what i expect, monitoring device's sleep mode at 50nA device turns on and consuming 70mA while connecting to network consumption raises to 400mA i want to see all this on graph current vs time.
Regarding connecting things with uCurrent in different way to get a particular output i would prefer to spend some money and get a long term solution.
-
@msat
What's wrong with using a shunt with your HP?
I'm not an experienced developer, did you mean? what i understand, I can shunt a resistor to my HP3457A to read nano range? can i get those reading directly to my computer to log?
you were going to spend on some hardware and get a HPIB to USB instead
How this hardware can help me?
@JohnnyMalaria
What's wrong with the uCurrent?
What I understand, I have to switch modes between nA, uA and mA that is not possible in real time scenarios, for example what i expect, monitoring device's sleep mode at 50nA device turns on and consuming 70mA while connecting to network consumption raises to 400mA i want to see all this on graph current vs time.
Regarding connecting things with uCurrent in different way to get a particular output i would prefer to spend some money and get a long term solution.
That makes perfect sense (pardon the pun). That's what I'm trying to do, too.
Why don't you try something like this: https://www.maximintegrated.com/en/products/analog/amplifiers/MAX40204.html (https://www.maximintegrated.com/en/products/analog/amplifiers/MAX40204.html)
There's an evaluation kit: https://www.mouser.com/ProductDetail/maxim-integrated/max40204evkit/?qs=zW32dvEIR3sqk4Ni%2FGgCcg%3D%3D&countrycode=US¤cycode=USD (https://www.mouser.com/ProductDetail/maxim-integrated/max40204evkit/?qs=zW32dvEIR3sqk4Ni%2FGgCcg%3D%3D&countrycode=US¤cycode=USD)
I bought one just a couple of weeks ago. Unfortunately, it's not useful for ac signals like I have but is intended for battery charging/discharging applications.
Maxim has quite a range of these types of products. You could also consider a programmable instrumentation amplifier such as the PGA281. I've just got an evaluation kit for that, too, but haven't tested it yet. https://www.ti.com/tool/PGA281EVM (https://www.ti.com/tool/PGA281EVM)
-
you were going to spend on some hardware and get a HPIB to USB instead
How this hardware can help me?
[/quote]
The HPIB to USB interface allows you to receive data from your bench meter to your PC for logging purposes which you can then plot out if you so choose.
-
There is nothing wrong with the uCurrent Gold, but...
You are working on ESP therefore may need to measure sleep current / wakeup power consumption and uCurrent Gold is not good for it as you have to toggle ranges manually.
It has to be the CurrentRanger (google it) with auto function to toggle between mA and nA or uA IMHO
-
Attached is picture from HP3457A manual, does it mean it can't go below 303.0 uA (by Max reading 3.3.0 uA they mean lowest range and we can't read something like 0.1 nA) can someone advise some used tool can be bought at ebay like DMM or power supply to measure/log nA
-
Attached is picture from HP3457A manual, does it mean it can't go below 303.0 uA (by Max reading 3.3.0 uA they mean lowest range and we can't read something like 0.1 nA) can someone advise some used tool can be bought at ebay like DMM or power supply to measure/log nA
"Max reading" means, well, the maximum value you will be reading in that range. You wonder about minimum or resolution. The 3457A is a 6.5 digit instrument, hence in the 300uA range the resolution is 100pA (0.1nA). There will be some amount of noise however. You might want to look at the specified accuracy (even though you can't expect such an old instrument to meeting its specifications anymore) to get a rough estimate of how small a value you can sensibly measure. For the 3457A in its 300uA range its accuracy is specified as 0.02% of the reading plus 104 counts. So yeah, forget about pA :-DD
Further, I'd caution that a fairly slow DMM like the 3457A might not be the ideal choice to determine battery life of a MCU project. Besides the burden voltage issue, it reports only at most 1350 readings/s (3.5 digits, AZ off), at high precision (6.5 digits, AZ on, 10PLC) only 2.5 readings/s. Those numbers assume a fixed range though. If the MCU sleeps most of the time with very small current (say less than 10uA) and only occasionally wakes up briefly (less than 10ms), when it consumes a few mA, the readings you'll get from the 3457A won't reflect average power dissipation closely.
-
If you know that the MCU can enter only a small number of discrete power states and you can force the MCU to linger in each for an extended time (say, a few seconds), then you should be able to measure the current with fairly good precision in each (the uCurrent shunt helps to minimize the burden voltage).
If you then monitor the current (via a shunt or the uCurrent) using a fast, not necessarily precise, instrument (say an oscilloscope) which allows you to determine the time spent in each power state, while your application is running, you'll be able to calculate the average power dissipation and hence battery life time with fairly good precision.
-
Not sure if I'm doing something goof, as datasheet of HP3457A says it can read minimum 303 uA but when i was working with my test circuit i can read
000.054 uA is it 54 nA? or my understanding is wrong (Please refer to attached picture)
however what i expecting from the load that was around this (50 nA) much current, I also verified it by putting additional load and removing it, I'm getting correct reading when i add additional load.
-
Your photo shows 54nA.
Where does it say that 303uA is the minimum. 303uA is the lowest (minimum) range, not the lowest reading. You can certainly measure to the nearest whole nA.
-
Where does it say that 303uA is the minimum. 303uA is the lowest (minimum) range, not the lowest reading.
Sorry i could not get you, by 303uA what i understand is, it can read 303uA current but if current is below 303uA like 200uA that's out of range
-
No, it means that the DMM has multiple ranges and the lowest range is 303uA and that range can measure up to 303uA. You can measure down to 1nA on that range.
It's like having two rulers - a meter long and 10cm long. Both have 100 divisions. So you can read to 1mm with the meter ruler and to 0.1mm with the 10cm ruler. BUT - if you want to measure to with a precision of 0.1mm, you are limited to a maximum of 10cm.
So, if you want to measure with a precision of 1nA, the largest current you can measure is 303uA. A larger current would require selecting a different range with less precision.