Calculating resistance

[Cantafford]

Hello,

I'm trying to make(in simulation not hardware) an analog to digital current converter. I'm using a LM134 current source: datasheet here.
I want my current source to output a current that can be set in the range: 2uA-100uA. On the "Calculating RSET" section from the datasheet it gives the formula. I looked at the electrical characteristics and put the parameters in the formula for RSET and arrived at a value of 13k ohms(I took the voltage as 1.2 volts)
Here's the RSET section of the datasheet:

Please note that this is my first time trying to calculate a value based on a datasheet ever so maybe there are noobies mistakes that I made. I would like if possible someone to tell me if I have calculated the value of RSET correctly and if with that value I will have a current in that range(2uA-100uA). Thank you for reading!

 

[spec]

Hi Cantafford,

It's been about three weeks since you posted your question about the LM134 current setting resistor, but if you are still interested here is my response:

I can remember as a noobie grappling with my first data sheet. It was for a Mullard EL84 tube and the first semiconductor data sheet was for a Texas Instruments 2S104 transistor. Both were pretty daunting at the time, but with a bit of practice you soon get the hang of data sheet language. And most importantly, you soon spot any errors and omissions. For example, the introduction to LM134 on the data sheet states, 'The sense voltage used to establish operating current in the LM134 is 64mV at 25°C' and, 'Zero drift operation can be obtained by adding one extra resistor and a diode.' I may be wrong, but from the formula in the data sheet, which is misleading, I make the first 67.7mV and the second is simply impossible, although much of the drift may be cancelled by the circuit shown.

Even though the datasheet is pretty comprehensive, you have chosen a challenging component in the LM134. It's a clever circuit but that makes it slightly difficult to understand. Also, it seems to me that Texas Instruments couldn't quite make up their minds if this device is a constant current source or a temperature sensor, and there are so many variables that it would be difficult to design a digital to analog converter with absolute accuracy using LM134s. That is not to say, though, that it is not a well-designed and very useful device.

By assuming the LM134 die is at ambient temperature and taking nominal values for the other variables you can derive a couple of formulas that allow Iset and Rset to be calculated easily.

The LM134's primary function is to produce a constant voltage between pins R and V-, ie Vr. By connecting a resistor, Rset, across Vr, Rset draws a constant current, Ir. In addition to Ir, the data sheet states that the LM 134 circuit needs around Ir/18 to function, so the total current, Iset = Ir+(Ir/18).

The actual Ir is found by, (227µV*°K)/Rset. The data sheet gives a misleading representation of, (227µV /°K)/Rset, which implies that 227ΩµV is divided by °K.

This formula can be further simplified by deriving the actual constant voltage from, Vr= 227ΩµV*°K. The normal ambient temperature shown in data sheets is 25 °C, which translates to 298.2°K. Thus Vr= 227µV*298.2= 67.7mV (the introduction to the data sheet states 64mV).

Thus, the simplified formula for calculating Ir is, Ir= 67.7mv/Rset.

Iset= Ir + (Ir/18), which works out to, Iset= 1.056*Ir.

This can be substituted to give Iset= (67.7mV*1.056)/Rset which works out to, 71.49/Rset.

So, finally, here is the practical formula for Iset:

(1) Iset= 71.49/Rset (where Iset is in µA and Rset is in KΩ)

And transposing to get a practical formula for Rset:

(2) Rset= 71.49/Iset (where Rset is in KΩ and Iset is in µA)

To answer your questions:

From your post, I think you say you make Vr= 1.2V, but my interpretation above of the data sheet gives a Vr of 67.7mV

From formula (1), an Rset of 13KΩ will give an Iset of, 71.49/13= 5.50 µA

From formula (2), you can get an Iset of 2 µA with an Rset of, 71.49/2= 35.745 KΩ.

Similarly, you can get an Iset of 100 µA with an Rset of, 0.7149 KΩ, or 714.9 Ω

I hope this answers your questions. Please note though that I have only been referring to the data sheet* for the LM134 and have never used this particular device in a design. Also the actual die of the LM134 has been assumed to be at a constant 25°C and the many other variables have been ignored. A worst-case toleranced design with the LM134 would be fun!

* Texas Instruments SNVS746E–MARCH 2000–REVISED MAY 2013 https://www.ti.com/lit/ds/symlink/lm134.pdf

 

[atferrari]

Just in case, the LM324 is an old quad op amp.

 

[spec]

Thanks atferrari: have corrected