Mohammad Alsmadi
-
Posts
16 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Gallery
Posts posted by Mohammad Alsmadi
-
-
Thank you @attila,
Just to confirm, if I use oversampling, the resolution is 16 bits. Then, the ADC resolution for the 50V range is around 0.7mV?
-
Thank you @attila,
The voltage range that I am looking to sense is ~4V. If I understand correctly, I should scale the signal down by 10 times using 0.1X attenuation and a range less than or equal to 200mV so that I can keep the resolution at 130uV, correct?
-
Hello,
The resolution for the ADP345 is 14-bit with +/-25V (50V difference) input range. Is it possible to adjust the input voltage range (reduce it) to achieve better voltage sensing resolution?
Thanks,
-
Thank you, JColvin, for your answer. By using the record mode, I cannot see the acquired values on the screen until the recording is done, correct? Is it possible to see the values as they change while using record mode?
Mohammad
-
Hello,
I am using the ADP3450. I am wondering if there is any way I can start sampling at time=0sec or time=0min. I tried to manipulate the parameters in the "Time" window, but I could not achieve what I was looking for. As shown in the attached photo, the sampling did not start at time=0. Also, when I try to export the raw data, for Ch#1, for example, it shows that the first data point is recorded at t=-147sec as shown in the second attached photo.
Could you please help?
Thanks,
Mohammad
-
Ok, John.
-
Hello John,
Thank you for your comment. I actually tried to use the power supply of the working to power the second one, however, the "device ready" LED is still off. Once I connect the USB2020 I hear the sound as if a new device is connected to the PC. However, I do not have access to the device.
Serial number for the first USB-2020 (good board): HE1793938
Serial number for the second USB-2020 (not working): HE1631214
Regards,
Mohammad -
Hello,
I am using USB2020. I have two devices. The first one was detected directly by the host PC and the "device ready" blue LED is ON. However, LED in the second one is OFF all the time. I made sure that the 9V input is present at the board input supply terminals. Also, I unplugged and plugged back the USB cable to the host PC. As shown in the attached photo, the device is present in the PC's devices list. However, I have no access to it. Also, I tried to see if Matlab toolbox can detect but I did not. Could you please help with that?
Appreciated,
Mohammad
-
Hey Arthur,
Thank you for your valuable comment. I already have the training data set to train my neural network on. Once I train the neural network, I believe I can save it such that I use it every time I run the script (So I do not have to train it every time). Do you think that is correct?
Coming to the performance part, could you please explain more? Do you mean the rate of the process when running the python script locally is slower than that of an average PC? I would appreciate it if you can provide me with more insights on the performance when running the python script with AI locally vs a PC.
Thanks again,
Mohammad
-
Hello,
I am planning to write my own python script using the SDK tool in Waveforms while using ADP3540/ADP3520.
Can I run a neural network in the python script? If yes, can the neural network be trained using an external data set? I plan on the following:
1) Training the neural network on a dataset that is already available.
2) Make the ADP device sample data (For example, a voltage value).
3) The sampled data is fed into the pre-trained neural network in real time.
4) The neural network gives an output based on the supplied sampled.
5) Sample data again.
6) A new output from the neural network.
7) repeat steps 2 to 6, and so on.
Is what I described possible? what could be the limitations?
I would highly appreciate your valuable answers and clarifications.
Thank you so much.
Mohammad
-
Hello John,
Could you please help me with the above issue? Thanks a lot.
Mohammad
-
Hello Joh,
Thank you for your reply. Then, I might try to use Labview to have Burst mode sampling and then acquire the samples in Matlab from Labview. Do you think this is a valid option?
Thank you!
Mohammad
-
On 11/11/2022 at 2:25 PM, JRys said:
Hello,
The USB-202 does not support BurstMode. The channel-to-channel time will be depend on the sample rate. For example, two channels @ 50k S/s will be spaced apart by 10uS, at 25k S/s it's 20uS and so on. BurstMode is available on the USB-1608G but only through our programming interface for C and Python.
The USB-202 does not support programmable ranges, it is fixed at +/- 10 volts. If you have small signals, the USB-202 may prove inadequate. In this case, the USB-231 isn't much more and is 16-bits over +/- 10 volts.
Best regards,
JohnHello John,
Thank you for your reply. It seems that you answered my questions based on the USB-202 not based on the US-B2020. Could you please help me addressing my questions based on the USB-2020?
Thank you for your time.
Mohammad
-
Hello,
I am writing with some questions about using Matlab to log data from MCC-USB 2020. I would appreciate your help.
1. Is it possible to use BURST mode in Matlab? If yes, is there any guide/documentation related to this?
2. Is it possible to change the input voltage range in Matlab? I assume that Matlab considers the +-10V range but would like to decrease this range if possible.
Thank you for your time and consideration.
Best,
Mohammad
Running Neural Network on Digilent ADP3450 in Linux Mode
in Test and Measurement
Posted
Hello
I understood that running a python script with a neural network on the ADP3450 is possible. Is the below doable?
1. Training a neural network to run in python script. The training can be done on a host PC (Offline).
2. Saving the trained neural network. When the neural network is trained, it has parameters that will use to predict the output when it receives a new input. I would assume the trained neural network to be saved on the Digilent.
3. Sampling waveforms in realtime in linux mode.
4.Running the trained neural network in a python script on the ADP3450 in linux mode and feed it with some information from the sampled waveforms.
5. Repeating 3 and 4.