voltage to current converter

On a sunny day (Fri, 16 Nov 2007 12:35:18 -0800 (PST)) it happened john wrote in :

None basically.

I would like to understand what exactly you mean by 'DC leakage'. In a previous post I showed you need no capacitor,

So then for a 0 to +ma output current you need -2.5 V on the other input of the INA.

No output cap either, but your load impedance must be low enough so the requested current can flow. If yo uhave a supply of plus and minus 10V, and a 5 k series resistor, then whatever yo udo,m the output current cannot be more thne +or- 10 / 5000. If your load is say 1K, then the output current can never be more then

+or- 10 / (5000 + 1000). So this limits the input voltage range, as I out is U1-U2 / 5000 and can never be greater then Iout max.

Using the 1M makes no sense as for 1mA you need 1000V.

So, practially the calc goes a bit like this:

U1-U2 = 2.5V, R=5000, Iout = 500uA, for 10V supply Rload total should be

10 / .0005 = 20000, you already have 5000 for R, so Rload must be < 15000 Ohm.

If I did not make any errors, but thisis the system.

What is your Rload?

Hi,

The DC leakage current will go into the load without the capacitors. The -2.5 volts offset is only there at the power up not after the DAC outputs the waveform. After the DAC outputs the waveform, the voltage at the non inverting input is zero so leakage gets low but still its there. I think that DAC zero and the ground zero has some offset and need trimming but can not do trimming because of PCB issues.

John

On a sunny day (Mon, 19 Nov 2007 12:28:58 -0800 (PST)) it happened john wrote in :

At power up the caps have charge zero, so the ywill not help, but perhaps even increase startup surges / currents in teh load.

3rd time I ask now: What is your load impedance.

OK I give up.

Hi,

The laod is in the range of 10kohm to 200kohm.

John

Hi,

The resistor in parallel with the current source will not let the capacitor charged upto +/- 15 volts. so when the power is turned off , the capacitor will have less charge to dump in the load. What do you think?

Regards John

On a sunny day (Mon, 19 Nov 2007 12:53:26 -0800 (PST)) it happened john wrote in :

OK, now think for a minute.

The output was (U1-U2) / R, where R was 5000. Your DAC does 2.5 V or more ful lsignal. the output current is then 2.5 / 5000 = .5 mA .5 mA in 200.000 Ohm = 100 Volt. Your supply voltage is? 10V? Go back to start and pay 100$ :-)

On a sunny day (Mon, 19 Nov 2007 13:07:51 -0800 (PST)) it happened john wrote in :

I was thinking about the other capacitor, it is a short at power up. But I was also thinking that none of this can work because see my other post.

Hello,

Would you please confirm two more things

1. What is the output impedance of this voltage to current converter? Ideally current sources have infinite output impedance.

1. If the voltage across the load is 1000 volts and power supply is

10volts. What would likely to happen a. the current will not flow b. the current will flow

Regards John

On a sunny day (Tue, 20 Nov 2007 10:39:06 -0800 (PST)) it happened john wrote in :

Ideally, yes. The output impedance is set by the open loop gain.

The current CANNOT flow, as you need 100V for it to flow. As you only have a much lower supply vltage, only a lower current will flow, resulting in the amplifier hanging one way against the supply.

On a sunny day (Tue, 20 Nov 2007 10:39:06 -0800 (PST)) it happened john wrote in :

Ideally, yes. The output impedance is set by the open loop gain.

The current CANNOT flow, as you need 100V for it to flow. As you only have a much lower supply vltage, only a lower current will flow, resulting in the amplifier hanging one way against the supply.

Jan Panteltje snipped-for-privacy@yahoo.com posted to sci.electronics.design:

Gosh, a fixed value source versus a controlled source but no difference? That does not make sense to me. Spice has controlled sources.

john snipped-for-privacy@hotmail.com posted to sci.electronics.design:

To bad the link got truncated. It makes it hard to figure out what you are trying.

On a sunny day (Wed, 21 Nov 2007 04:11:27 -0800) it happened JosephKK wrote in :

look at this in context, he has a controlled source. Any constant current source can be controlled in some way, If you really wanted to be pedantic.

F*ck spice. Reality rules.

Jan Panteltje snipped-for-privacy@yahoo.com posted to sci.electronics.design:

:

It seems unlike you to not follow the chain properly.

On a sunny day (Fri, 23 Nov 2007 04:48:24 GMT) it happened JosephKK wrote in :

Reality rules.