Electronics > Beginners

DC dummy load circuit calibration

<< < (18/20) > >>

VEGETA:
Here is latest simulation after adding 2k resistors to TL431 reference + my new circuitry mentioned above (LM358).

I claim that this one solves the RF problem or at least make it less since now instead of 12uA current through our 10-turn POT (which is mounted on front panel), we have around 0.2mA which is more or less same as other designs including Dave's (5v on 10K but I have 2.5v on 10K).

I made 52R to standard 47R then re-adjusted zeroing POT to make it exactly 1V per 1A.

Looking for Ian's final opinion on this.

Ian.M:
I'll try and take a proper look at it later today.
First impressions:
OPAMP U2 isn't needed - take the 1K:47R divider output straight to in+ of the current sink feedback OPAMPs

Since you've now got a spare OPAMP, it would be best used in an inverting configuration to generate a precise and stable -0.5V rail for the zero adjust.   Use 2x 10K in parallel from the 2.5V reference and 1K for feedback.

Edit: I've now had a chance to look at it:
I've reworked it, to improve the simulation, and have also used the second OPAMP in the LM358 to generate the zero adjust voltage.  Hopefully I've included all the required symbols and models with the sim.

I know you aren't particularly fond of my potentiometerVC symbol, however it really is the easiest way of #1 showing the part is a potentiometer or preset, not two fixed resistors, and #2 being able to sweep or step the wiper position, either for a .dc sim or with respect to time in a .tran sim.

VEGETA:
Here is your modification, I did it quickly though. I tried it on 1.000A and it can be achieved but other values don't work with the same calibration.

Ian.M:
I've done a lot of tweaking of the sim this morning to get the full output range, good zeroing, and also to save supply current e.g. the LED + resistor now feeds the TL431 shunt regulator, as that needs about 10mA to regulate well.

See attachments to reply #86.   You may have to change some resistor values to get a suitable range for the zero adjust - see notes on schematic.


By adding one SPDT switch, an input socket, and a 1K resistor and a pair of diodes for input protection,  you can control the load from a function generator, or a low pass filtered Arduino PWM which opens up possibilities like automated discharge tests for batteries with datalogging or using it as a constant power or constant resistance load if you monitor the load voltage on an ADC channel.

VEGETA:

--- Quote from: Ian.M on June 04, 2018, 12:11:33 pm ---I've done a lot of tweaking of the sim this morning to get the full output range, good zeroing, and also to save supply current e.g. the LED + resistor now feeds the TL431 shunt regulator, as that needs about 10mA to regulate well.

See attachments to reply #86.   You may have to change some resistor values to get a suitable range for the zero adjust - see notes on schematic.


By adding one SPDT switch, an input socket, and a 1K resistor and a pair of diodes for input protection,  you can control the load from a function generator, or a low pass filtered Arduino PWM which opens up possibilities like automated discharge tests for batteries with datalogging or using it as a constant power or constant resistance load if you monitor the load voltage on an ADC channel.

--- End quote ---

Good job re-arranging it, it is better now.

#############
# your POT model #
#############

I don't hate it but I don't understand it. Now it is obvious to me that it relies on voltage to change resistance. I wish to understand how it is made.

###########
# Zeroing POTs #
###########

I don't quite understand R1 and R3, their values are standard but how they are mixed with RV2 and RV3?

What I understand is that RV2 sets the negative offset via op-amp U6 which acts as an inverting op-amp. So it takes the gain set by R1 and R3 and inverts the signal coming from RV2 according to it... is this correct?

But here RV2 still what truly changes the result so R1 and R3 are a fixed maximum limit to the offset?

Also, RV3 is a bit vague to me, you say it is for making 1V per 1A. I see a divider formed by R10, R11, and RV3 while RV3 is outside op-amp U6 loop. You seem to put it as 100 ohms potentiometer so it is not the same as 10K pots used elsewhere which means I have to get 100R ones just for this.

This is the one: https://lcsc.com/product-detail/Precision-Potentiometer_BOURNS_3296W-1-101LF_100R_C83688.html

############
# TL431 + diode #
############

Will 10mA be enough for this to work? I mean for the LED it is but for 2.5v shunt regulator output? especially that I would use 2 of 1K in parallel to get 500R instead of 470R since the later one is not used but here.

I didn't quite understand D_TL437 thing as it seems to be yet another one of your special models. First, I thought you would put a zener diode as a model for TL431 since it doesn't have an official model. It won't matter much here since it is just a voltage source.

###############
# Function generator #
###############

I guess I will put an opening for it in the back or somewhere not visible since I won't be using it.

If someone put more than 2v it would be bad. As for protection diodes, I thought of putting 1N4001/7 since they are available to me and others. As I understand its operation, if input voltage from function generator (or pot for some reason) exceeds 2.5v by 0.7v then diode will conduct. At the same time, it acts as a reverse protection.

But this won't prevent someone from putting 3v or so as an input, then I wonder how the circuit will react.




##########

I will try to update the schematic in KiCAD probably tomorrow with this new update. I guess now we won't worry about RF issue anymore which is a big plus.



Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod