+/-3.5v to 0-100k resistance conversion circuit help please!

[ljcox]

I need more time to do the calculations. Hope to post in an hour or 2.

 

[ljcox]

Here it is. It is based on Ron's level shifter and current mirror circuits.

I could not find any data for the NTE930. Do you have a data sheet?

The CA3130 is available here, so I'm surprised you can't buy them. A dual version of the CA3130 would save space - if you can find one that will operate on 5 Volt and has a common mode range down to at least 0 Volt. The frequency response if not critical.

Does anyone have a suggestion?

Edit, I have updated the attachment - forgot to include a 100 nF bypass capacitor. After 5 attempts, hopefully it will work this time.

 

[djpubba]

Here it is. It is based on Ron's level shifter and current mirror circuits.

I could not find any data for the NTE930.

Nobody's got it in stock around here anyway. How about a NTE858?

https://www.nteinc.com/specs/800to899/NTE858M.html

What kind of transistor is Q2? Also is R7 really a 200 ohm pot or should it be 200k?

Thanks,
Tim

 

[ljcox]

Q2 and Q3 are the same. This part is a copy of one of Ron's earlier posts and what you have already built and tested.

R7 is 200 Ohm, not k ohm! R3 is 270 ohm.

I looked at the data sheet link. But it does not say if this op amp will operate on 5 V. I'll try elsewhere to see if there is more info.

 

[ljcox]

I downloaded the NTE858 data sheet and it does not have enough common mode range and it will not (as far as I can gather) operate on 5 Volt.

You need an op amp that has a common mode range that includes 0 Volt and operates from 5 Volt.

Does anyone know of an equivalent to the CA3130 op amp?

The frequency response is not critical, just the parameters mentioned above.

 

[ljcox]

In case you cannot find an alternative IC, I will post a circuit later that uses the DC - DC converter ICL7660 (there is also a MAX1044) that will provide -5V from +5V. Thus you can use op amps such as

LF412, TL072 or TL082. These will all operate from +/-5V supplies and will suffice. They are all duals.

 

[ljcox]

Here is a reply from another thread. These won't need the 56 pF capacitor. I have not checked their pin arrangements.

LM324 quad or LM358 dual. (low power)
MC33171 single, MC33172 dual and MC33174 quad. (low power)
MC34071 single, MC34072 dual and MC34074 quad. (normal power)

They all work with a 5V supply and their input common mode range includes ground.

 

[Roff]

Djpubba, if you are pursuing my circuit, I think you're wasting your time. I spent some time today comparing an RC multivibrator with the voltage-controlled current source multivibrator, and I'm convinced that Len is correct. The current is indeed proportional to voltage, but the pulse width is inversely proportional to current, so you wind up with a hyperbolic relationship between voltage and pulse width. Right now, I'm out of ideas.

 

[ljcox]

Djpubba, if you are pursuing my circuit, I think you're wasting your time. I spent some time today comparing an RC multivibrator with the voltage-controlled current source multivibrator, and I'm convinced that Len is correct. The current is indeed proportional to voltage, but the pulse width is inversely proportional to current, so you wind up with a hyperbolic relationship between voltage and pulse width. Right now, I'm out of ideas.

I've also spent the day checking the calculations of the circuit I posted earlier and considering alternatives.

Although the current should be roughly proportional to the inverse of the pot resistance, the calculated currents do not agree with your measurements. So I don't think this circuit is going to give you the results you want and I don't know why.

I'm sorry, but I can't afford to spend any more time on it.

Current required Current from circuit Screen
30 uA----------------30 uA-----------------999
60 uA----------------304 uA----------------500
???-------------------1.38 mA--------------000

??? means I can't see where you measured this, so I can't recall from where I derived the 1.38 mA, but that's the figure I used for the design.

So try it if you wish, but you may be disappointed.

 

[djpubba]

Thank you for all your brain hours of thought on this. I'm probably not going to suggest using an original Vectrex controller in this project as I originally intended.

FWIW, you can check out part 1 of the article here, if you like:

**broken link removed**

It's in issue 7, pages 30-32 (the 16th page of the PDF version).

Part 2 with what to do about controls will appear in the next issue (which was due today!).

 

[ljcox]

You're welcome.

I realised why the circuit did not work as expected. The fundamental problem is that the pot only has a 5.2:1 range as Ron pointed out.

In order to solve this issue, he suggested using a level shift. But the result was still not "linear" as viewed on the screen.

I then suggested we use a current amp. so the reciprocal of the pot resistance determines the current into the computer. But I also needed to level shift it in order to increase the range. So it is the level shifting that is modifying the reciprocal function so it is no longer a proper reciprocal.

I expect that a digital solution would work, but you would need a digital pot that has a resistance of 100k.