To HBB; that circuit will work fine, it is basically identical to the one I posted but using a high gain top transistor (made from 3 transistors), and it adds a pot to the lower transistor to allow current adjustment.

Personally I would eliminate the pot and just use a fixed current sense resistor (but make the resistor easy to change). The current limit will occur when the voltage on that resistor equals about 0.7v.
So; R = V/I
to set current of 1A; R = 0.7/1 = 0.7 ohms
to set current of 2A; R = 0.7/2 = 0.35 ohms

Note, with the 0.5 ohm resistor in your circuit the MINIMUM current limit will be about 1.4A (I = V/R = 0.7/0.5 = 1.4A), and the current adjustment will go UP from there.

To Tinamisra; Wow! I think you need a better translator program!

 

To RB,

Thanks for the details.
Yesterday I succeeded to make pump run in stable conditions very near the max flow point of the well, which is exactly 900 liter/hour.

I feeded the pump with 48 VDC from my batt. bank and the made two work points over 900 L/h that of course activated the dry condition cut-off device.
Then the first point below them gave this results:

Vpump 33.5 Vols (average of many digital readings)
Apump 3.5 Amps (average of many digital readings)
Ppump 117.25 Watts

Ptotal 50 V (average of batt. voltage) * 3.5 Amp = 175 Watt

Lost in series resistors 175-117.25 = 57,75 Watt

Volume pumped 1000 liter tank full in 70 min = 857 Liter/hour

The two other working points were below 857 liter/hour so no interest to work on.

One remark: the series resistors were formed by copper wire (old transformers unwound by me) + a tractor lamp of 12 V 60 Watt.
The lamp was continuously changing brightness, is a nice instant ammeter ! That shows that the MPPT logic is continuously working even been fed by a stable DC source.

Thanks for the comments about the BJT limiter. Of course I will use a fixed resistor for reliability.

I will limit current output @ 3.5 amp and problem solved.

After having this BJT pass transistor up and running, I will try a switching limiter to save some of the lost energy, but that will be a later step.

Have a nice week end,
Horacio

 

Excellent news and thanks for posting the info!

If you do want to save that 57 watts wasted in the constant current circuit (which by the way will need a MASSIVE heatsink) all you need now is a 33v 3.5A buck SMPS with both voltage and current limiting, or actually it just needs the 3.5A current limiting.

Unlike your original choice of a 90v 1A boost SMPS, a 33v 3.5A buck SMPS is a much much easier build.

 

Thanks a lot for all your help.

Tomorrow I will build the BJT pass limiter, and I will sacrifice 57 Watts vs time. That current limiter will be in service immediately.

Now, I am decided to learn and build the SMPS buck converter. I know the principle, but never saw one buck converter which is capable to limit its own current output.

Please guide me to where to start from.

Thanks again,
Horacio

 

I have been reading TI, Linear and Maxim websites. there is a lot of ICs that can do what I wont. I need Vin in the range of 58-46 Volt, Vout at 33 V, and current at 3.5 amps.

Linear has an IC selector module in their web.

I found the attached circuit that shows how the blue shaded limiter senses Out Current in a 20mOhm R, compares with the internal 1.2 VRef of the IC and injects a signal when current goes high that limits value to ISEN pin of the IC. This terminates the ON cycle and limits energy transfer.

The concept is either clear and elegant, but how to materialize it for my parameters?

Am I on the good track ?

Thanks, Horacio

 

Actually, sorry about to change direction again but I think you can do it another way!

You said you are getting the desired motor performance at;
Vpump 33.5 Vols (average of many digital readings)
Apump 3.5 Amps (average of many digital readings)
Ppump 117.25 Watts

So the MPPT is adjusting it's PWM to get the max possible watts at the motor of 117W.

If you set your constant current device to a lower current, the MPPT will increase voltage in, and you may be able to "tune" it to be similar efficiency to a SMPS device.

For example;
117W @ 48v = 2.44A. (I = P/V, = 117/48)

So if you set the simple constant current circuit to 2.44A the MPPT would just self adjust to give you as close to 117W as it can get. The current limiter will still drop a couple of volts, but even at 3v 2.44A now it is only wasting about 7 watts, so it's getting close or equal to the efficiency of a SMPS buck circuit.

Can you do a test with a 2.44A current limiter?

 

Hi RB

I am in my office in Buenos Aires now, till wednesday.

It is a clever test, I did not realize that before.

I will go to the "campo" in Entre Rios on Thursday and I will carry on resistors to build the limiter for many currents, from half Amp by half Amp.

I will post results as sonn as possible.

Have a nice day,
Horacio

 

Bingo !

I tuned the current limiter to 2.5 Amps and in that condition the MPPT is taking 45 Volts, the rest is dropped in the limiter.
So the motor is getting 112.5 watts which is perfect.

Thanks so much your all the received help !

Horacio

 

Wonderful!

After considering building a tricky 98v >100W boost converter it's great to see the problem solved with something as simple as a 2.5A constant current limiter.