# I find I keep asking the wrong question...

Discussion in 'General Electronics Chat' started by trs300, May 28, 2010.

1. ### trs300 Thread Starter Member

May 22, 2010
29
0
It's frustrating to know as little as I do about something... Ive asked this before but I think I asked it the wrong way... So I'll try again.

How do I buid an op amp circut, with an easily available op amp (frys or some place like that) where it is an adder circuit...

I want to use a 0-5V 20ma signal and translate that signal to a 9-14 volt power supply at 1.5A max,

So it's straight translation from the signal circuit to the power curcuit, we are adding 9V to the signal.

0-5v @ 20ma = 9-14V @ 1.5A.

It does not have to be exact. anything close will do... Assume I have sufficent power supply of 14.5V already limited at 1.5A max for the power circuit...

I've been trying to figure this out for a week. For someone not knowlwdgeable in all this I have to say my head has pretty much turned to mush... I have a lot of respect for you guys...

Thank You...

---

2. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
784
A power supply of 14.5V may not give sufficient headroom to regulate the output to 14V when the input control voltage is at its maximum of 5V. If you are using a linear control solution, you will probably need a couple of extra volts to allow for the main control element [e.g. power transistor] voltage drop and/or your op-amp dynamic output range limitations. That's assuming everything is powered by a common DC supply.

A more complex switch mode solution may allow the use of your pre-existing 14.5V DC source.

Last edited: May 28, 2010
3. ### Wendy Moderator

Mar 24, 2008
21,012
2,745
To add to what t_n_k is saying, many op amps can not go rail to rail, they can only get withing two volts or so of the power supply voltage. Can you provide more positive voltage for the op amp? I think I have a circuit that can increase the drive, though I've never tested it.

Use this as a high power output of an op amp circuit.

Creating a Virtual Power Supply Ground

Adjust R2 until the circuit draws around 5-10ma, this puts it into a linear mode, and use the output shown as ground as a straight output. Showing the circuit another way...

This circuit has no voltage gain, but huge amounts of current gain. Eliminate the input and output capacitors to bring it down to DC response, and R6/R7 can be reduced to increase the current output. They are needed to stabilize the currents in the circuit, so they can't be eliminated entirely.

Last edited: May 28, 2010
4. ### trs300 Thread Starter Member

May 22, 2010
29
0

That said... I can shift the control voltage upward... Instead of 0-5V it could be 3-8V. But it would still be at very low current. 20ma to 50ma maybe... Would that help things?

5. ### Wendy Moderator

Mar 24, 2008
21,012
2,745
It's not the control voltage, it is the output drive of the op amp. I suspect you were replying same time I was. For 14V output of the op amp you need 17V on the power supply, it is what we are referring to as headroom.

6. ### trs300 Thread Starter Member

May 22, 2010
29
0
Thank you for the lead Bill... It will take me some time to try and understand what I'm reading however... But I do appreciate the help.
--

7. ### Wendy Moderator

Mar 24, 2008
21,012
2,745
Op amps don't generally have much drive, the conversion is easy enough, but not with current output. Some do, but they aren't going to be easy to get..

The problem is the power supply feeding the op amp. If you want 14 volts out of it you need power supply voltage enough to have the op amp reach the desired output voltage.

8. ### trs300 Thread Starter Member

May 22, 2010
29
0
OK... That makes more sense... It may actually be ok if it's a few volt's low... It's hard to explain. But this circuit is supplimenting another curcuit so that the other circuit is getting more current sooner then it would normally. It's like a bypass on the same line bypassing the normal controller, which is throttles back to much in the midrange. But under maximum duty, the normal curcuit will be supplying the full voltage available anyway... and this supplimental curcuit (bypass) will have less effect. The bypass has more purpose in the lower and mid ranges of control...

To make it simple. If I couldnot get the full 14V on the top side might be ok... Because the normal curcuit will supply to top end (all that there is) anyway. I could try it with what I can get, and that might work just fine...

I'm a mechanical person. So let me explain it the way I know how... I have a 0-5V control signal that is the same as full closed to full open... 0-100%. I want that signal to operate a valve that throttles current. At 5V control signal I want the current valve open wide with as little restriction as possible. at 0v control signal I want the current lave to cut back on as much current as possible... The flow we are controlling is a 14 to 14.5V source with a 1.5A potential flow rate...

---

--

Last edited: May 29, 2010
9. ### SgtWookie Expert

Jul 17, 2007
22,194
1,764
Bill,
Those circuits you posted are not only fairly complicated for a n00b, but will not be stable nor reliable over the automotive temperature range.

10. ### Wendy Moderator

Mar 24, 2008
21,012
2,745
I just reviewed the OPs post, where are you getting automotive in this?

11. ### SgtWookie Expert

Jul 17, 2007
22,194
1,764
OK, you are kind of coming into his threads new.

It is an automotive application for a shift control point.

Anything that is used in a circuit he wants must be rated for automotive temperature ranges. He doesn't understand why that is necessary yet.

12. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
784
So you'd kind of hope that trs300 would have enlightened the rest of us - mind reading certainly isn't my speciality nor presumably Bill_Marsden's who went to a lot of effort.....

13. ### Wendy Moderator

Mar 24, 2008
21,012
2,745
Naw, I just pulled stuff from my library. Automotive matters though, since it is the next step down from milspec.

14. ### t_n_k AAC Fanatic!

Mar 6, 2009
5,448
784
OK - I was ignorant of that fact. Thanks!

SgtWookie - ever vigilant!

15. ### trs300 Thread Starter Member

May 22, 2010
29
0
Wanted to take a moment to thank everyone for their help. The recommendations here helped me a lot in how to think about ways to get the result I was after. As for the automotive comments. I'll admit I'm still not clear on what the issue is there... The circuit will sit under my dash and be seeing the same climate that I will? I just do don't see this concern for "severe conditions" that would be seen by the components...

Anyway, Oddly enough the answer was under my nose all week. The LM317T does exactly what I needed... I dawned on me over the weekend that the configuration of the two resistors on this regulator were actually setting a positive voltage on the adjustment leg, that this adjustment leg senses voltage not current, and that the voltage sensed on this leg directly controls the amount of current that moves through the semi-conductor. Once I realized this I simply removed the resistors and fed my MAP signal voltage to the adjustment leg and Bingo. It regulates a 1.5 amp current from around 1.2V to 12.5V from a 14V supply. The Vout voltage stays to within 1V of the voltage sensed by the adjustment leg. I did have to supply 14V to the Map sensor to do this, so I may still need a low current op amp to allow me to drop the MAP sensor back down to 5V and and amplify the MAPs' signal so that I can use it on the adjustment leg of the LM317T...

I'm just happy to have found a simple and readily available solution to this requirement... Turned out to be the "Use something in an un-intended way" type of solution...

Thanks again...

--

Last edited: Jun 1, 2010
16. ### SgtWookie Expert

Jul 17, 2007
22,194
1,764
trs300,
The LM317 is not rated for operation over the automotive temperature range.

If you are sourcing 1.5A current from the regulator with the MAP output at minimum, you will be dissipating approximately (14v-1.25v)*1.5A = 12.75*1.5 = 19.125 Watts of power in the regulator. You will probably not be able to remove the heat fast enough to prevent the regulator to go into thermal shutdown at 125°C, unless you have it immersed in a bath of liquid nitrogen.

The LM317s' absolute maximum input to output voltage differential is 40v. You may have spikes exceeding 60v during a load dump, such as when you turn off the headlamps, or you release the brake pedal and the brake lights switch off.

You will most likely fry the MAP sensor within the first few times of turning off your headlamps due to the load dump.