@@uni-byte I first tried cross wiring a brake relay to the remaining charge on the Capacitor vgave it a reverse pull. That was hard on the commutation brushes but it stopped the motor quick . I changed that to a simple short through N.C. contacts. 1 /4 turn stop but hard on brushes. Used a 7805 with a 2 ohm resistor as a constant current 2.5 amp regulator,and a 48 ohm resistor in parallel. The motor was 24 voltdc rated 3 amps. The motor stopped little over 2/3 of a turn. The motor drove a score cam motor which synced up switches. It had to land on a position within 2 * degree. That gave it a constant brake . It had a gearbox of 76 to 1 so it actually worked.
I would have been tempted to design this with power resistors, and using MOSFETs only in full-on or full off mode to switch the resistors on and off, like a relay really. Of course you don't get fine-grained control of the load, but you get a step-wise load depending on how many resistors you turn on. I like your approach better, overall.
That would be an interesting approach. Actually I kind of had that set-up some years back. I had a whole load of 200W power resistors (1, 2, 3, 5, 10 & 15 ohms) and put them in series or parallel as required using HD alligator clip wires. I've even done it recently.
Would it oscilate at the temp threshold, or would it have protection hysteresis/push to reset? Or is this for now only a prototype to test if it is viable in principle? Also, have you thought about watercooling, or using something with high thermal mass? Also x2, there are aplenty of 100-150W CPU fan+radiator units that are pretty cheap and efficient, or is "bulkiness" an important factor?
The temperature threshold will need a power recycle to clear it. I want it that way on purpose so that I have a chance to clear/investigate what caused it. However, the whole thing is really an experiment. If I get to 100W that'd be golden, but 75W will more than meet my needs. To your last question, yes, bulkiness is an issue as are long leads going to the MOSFET. The former because I have a cabinet in mind to meet the form factor I want and the latter to prevent instability. They both kind of play hand in hand for this particular build although almost everything would definitely be easier to accomplish if I had more space. Maybe for version II, if I need it.
Interesting project, im personally interested in. Some thoughts: you can still use only one Opamp if you use a good Linear L2 FET. IXYS still has some (e.g. IXTN90N25L2) in a heavy SOT227B package. What I'm also interested in is to have dynamic load feature to provide a load step for power supply testing (stability). How is the load finall controlled (petentiometer or microcontroller)?
Hmm, On teh L2 MOSFETs, maybe but since the LM324 is only $0.40 from Digikey, why not have it in there for better control and stability? On the step performance, I really don't know, but my guess is with the 4K7 resistors and the 0.1 caps it going to change current rather leisurely. I guess those could be tightened up a bit to react quicker by pushing the stability a bit. I have the SIglent load for that sort of stuff, so that was not my main goal in making this one. For my use the control will be via coarse/fine potentiometer. You could use an MCU or small computer. It just needs a 0.00V to 0.100 volt control voltage.
I'm sure you are aware, but it may be useful to explain to viewers how the limits of temperature in C need to be translated into K (temperature rise) in order to allow use at different ambients. In my world (IEC61439) we test at whatever ambient and the limits are all in K to make it easy. This looks like a great project for testing batteries as well as power supplies.
The IEC have adopted the Kelvin scale as they deal with much wider temperature ranges throughout all their standards but in the electronics side of EE The Celsius scale is more usual. There are a few areas where using K instead of C is advantageous, like in understanding the physical response of thermocouples and thermistors, but even then we are usually only interested in their response over a specific range, so C is generally good enough for most electronics work.
I agree batteries and power supply testing would be my primary use case but I don't get the temperature bit, all will become clear when it's switched on and smoke tested
So, I'm likely going to limit it to 20A max (mostly because of the traces on teh PCB). In that case each resistor would need to pass 5A each (regardless of wattage) maximum. The power dissipated by those resistors would then be R times I squared or 5x5x0.02 = 0.5W. They should be fine. The heavy lifting will be done by the MOSFETs.
![SahBabii - Viking [Official Music Video]](http://i.ytimg.com/vi/D37cL92JX0I/mqdefault.jpg)
Yes! Finally a DC load with a control circuit per MOSFET! 👍
Yes indeed!
cool!
Thamks!
Awesome, this is something I need so I will glued to the project, can't wait ! I also will value low voltage stuff too.
Good to have you aboard Andy!
I use a similar circuit to evaluate PV cells and graph their current / voltage curves. Great project! 73
It's good to hear that someone else is building these circuits.
Looks like a fun and really useful project.
It's been fun in my mind anyway. Let's hope the reality is as good! LOL!
I used a 7805 as a dynamic brake for a small motor.
Cool idea. Did it work well?
@@uni-byte I first tried cross wiring a brake relay to the remaining charge on the Capacitor vgave it a reverse pull. That was hard on the commutation brushes but it stopped the motor quick .
I changed that to a simple short through N.C. contacts.
1 /4 turn stop but hard on brushes. Used a 7805 with a 2 ohm resistor as a constant current 2.5 amp regulator,and a 48 ohm resistor in parallel.
The motor was 24 voltdc rated 3 amps. The motor stopped little over 2/3 of a turn. The motor drove a score cam motor which synced up switches. It had to land on a position within 2 * degree.
That gave it a constant brake .
It had a gearbox of 76 to 1 so it actually worked.
@@allanpatterson7653 Very nice!
I would have been tempted to design this with power resistors, and using MOSFETs only in full-on or full off mode to switch the resistors on and off, like a relay really. Of course you don't get fine-grained control of the load, but you get a step-wise load depending on how many resistors you turn on. I like your approach better, overall.
That would be an interesting approach. Actually I kind of had that set-up some years back. I had a whole load of 200W power resistors (1, 2, 3, 5, 10 & 15 ohms) and put them in series or parallel as required using HD alligator clip wires. I've even done it recently.
@@uni-byte Another method might be a PWM-based variable load. A lot of ways to skin a cat, I suppose.
@@yeroca Yep!
👍👍👍👍👍
👍
Would it oscilate at the temp threshold, or would it have protection hysteresis/push to reset? Or is this for now only a prototype to test if it is viable in principle?
Also, have you thought about watercooling, or using something with high thermal mass?
Also x2, there are aplenty of 100-150W CPU fan+radiator units that are pretty cheap and efficient, or is "bulkiness" an important factor?
The temperature threshold will need a power recycle to clear it. I want it that way on purpose so that I have a chance to clear/investigate what caused it. However, the whole thing is really an experiment. If I get to 100W that'd be golden, but 75W will more than meet my needs. To your last question, yes, bulkiness is an issue as are long leads going to the MOSFET. The former because I have a cabinet in mind to meet the form factor I want and the latter to prevent instability. They both kind of play hand in hand for this particular build although almost everything would definitely be easier to accomplish if I had more space. Maybe for version II, if I need it.
Interesting project, im personally interested in. Some thoughts: you can still use only one Opamp if you use a good Linear L2 FET. IXYS still has some (e.g. IXTN90N25L2) in a heavy SOT227B package. What I'm also interested in is to have dynamic load feature to provide a load step for power supply testing (stability). How is the load finall controlled (petentiometer or microcontroller)?
Hmm, On teh L2 MOSFETs, maybe but since the LM324 is only $0.40 from Digikey, why not have it in there for better control and stability? On the step performance, I really don't know, but my guess is with the 4K7 resistors and the 0.1 caps it going to change current rather leisurely. I guess those could be tightened up a bit to react quicker by pushing the stability a bit. I have the SIglent load for that sort of stuff, so that was not my main goal in making this one. For my use the control will be via coarse/fine potentiometer. You could use an MCU or small computer. It just needs a 0.00V to 0.100 volt control voltage.
I'm sure you are aware, but it may be useful to explain to viewers how the limits of temperature in C need to be translated into K (temperature rise) in order to allow use at different ambients. In my world (IEC61439) we test at whatever ambient and the limits are all in K to make it easy. This looks like a great project for testing batteries as well as power supplies.
I don’t follow.
The IEC have adopted the Kelvin scale as they deal with much wider temperature ranges throughout all their standards but in the electronics side of EE The Celsius scale is more usual. There are a few areas where using K instead of C is advantageous, like in understanding the physical response of thermocouples and thermistors, but even then we are usually only interested in their response over a specific range, so C is generally good enough for most electronics work.
I agree batteries and power supply testing would be my primary use case but I don't get the temperature bit, all will become clear when it's switched on and smoke tested
@@andymouse Well, let's hope there isn't too much smoke!
@@uni-byte :)
With only 5w resistors I don't know if you will be able to push that load to 100w
So, I'm likely going to limit it to 20A max (mostly because of the traces on teh PCB). In that case each resistor would need to pass 5A each (regardless of wattage) maximum. The power dissipated by those resistors would then be R times I squared or 5x5x0.02 = 0.5W. They should be fine. The heavy lifting will be done by the MOSFETs.
@@uni-byte :)