LM234 current source problems?

[Flyback]

Hello,

I wish to make a 6.5mA current source with LM234.

The ratio of I(bias) to I(set) isn't quoted in the datasheet for any more than 5mA, therefore, i must use two pieces of LM234 and parallel them. (3.25mA each)

This means using a resistor of 21.2 Ohms in each LM234 circuit.

1....Is it ok to parallel LM234's

2...The datasheet page 3 quotes the ratio of I(set) to I(bias) as 14 for Iset= 1mA to 5mA, ...However, the graph on page 5 of the datasheet clearly shows that this ratio varies widely over the range 1mA to 5mA.....so how come they have quoted a fixed figure of "14" on page 3 of the datasheet?

3.....Why does the datasheet indicate that LM234 can supply 10mA?....after all the ratio of I(bias) to I(set) isnt stated for any more than 5mA....so surely 5mA is the realistic maximum for LM234?

LM234 DATASHEET:
https://www.electro-tech-online.com/custompdfs/2013/05/lm234.pdf

 

[dougy83]

I wish to make a 6.5mA current source with LM234.
The ratio of I(bias) to I(set) isn't quoted in the datasheet for any more than 5mA, therefore, i must use two pieces of LM234 and parallel them. (3.25mA each)

The datasheet says you can go up to 10mA, 400mW Pdiss. It also says that any current over 100uA will affect the regulation due to self-heating.

You can just buffer the output current by connecting a PNP transistor as per figure 20. This not only increases the current capabilities, but also reduces the self-heating of the regulator.

1....Is it ok to parallel LM234's
Yes, in the manner you outlined with two separate resistors, you simply have two additive parallel current sources. An external buffer transistor also works.

3.....Why does the datasheet indicate that LM234 can supply 10mA?....after all the ratio of I(bias) to I(set) isnt stated for any more than 5mA....so surely 5mA is the realistic maximum for LM234?
The abs. maximum is 10mA, but it seems better suited for 100uA-500uA (based on the flatness of the Iset:Ibias curve and self-heating).

 

[Flyback]

Figure 20 is always going to have the external PNP biased ON, because of the bias current flowing into the V+ terminal of LM234............so in that case, there will be a minimum current flow in that circuit....what would that minimum current value be?
...i imagine it would be >7mA because of the high gain of the PNP....................in other words, figure 20 would always flow at least 7mA....so if you wanted 6mA than fig 20 would be no good?

 

[dougy83]

The minimum current will be the bias current multiplied by the hfe (current gain) of the PNP. The transistor reduces the output current of the LM234 and therefore reduces the bias current as well.

e.g. for 6mA out, if hfe = 200, the lm234 output current will be ~30uA and the bias current will be ~2uA.

 

[Flyback]

...thanks, but LM234 will always draw more than 2uA of bias current, -to furnish its internal circuitry............so in other words, it would not be possible to choose a current of 6mA, as the bias current drawn would mean more than 6mA of current flowed.

 

[dougy83]

...thanks, but LM234 will always draw more than 2uA of bias current, -to furnish its internal circuitry............so in other words, it would not be possible to choose a current of 6mA, as the bias current drawn would mean more than 6mA of current flowed.

Good thinking, but no. Figure 12 shows Ibias to be 625nA for an Iset of 10uA.

With the PNP, the bias current will be a problem for low Iset currents, not for larger values of Iset. If you need a small output current or if you're using a transistor with high gain, it's very simple to place a resistor across its emitter-base pins to reduce and offset the current gain.