Odd L200C Behavior

Thread Starter

anw

Joined Jan 11, 2014
10
I'm building a three-channel, 0-28V, 0-2 amp (both variable) power supply. I took the schematic straight from the L200C datasheet and just triplicated it (common input with an added 12 V reg for a fan, besides that just three duplicated strings). Here's the design, identical to the app circuit in the datasheet (the right hand schematic):

Well, I can't seem to insert the schematic here, but anyone familiar with this chip will probably know, it's straight from the L200C datasheet, variable voltage with variable current limit. I'll keep trying; this is my first post and I've probably broken some rule. Meantime, here's a link to the picture, and it's the schematic on the right:
http://www.eeweb.com/blog/circuit_projects/l200-based-voltage-regulators

When I hook it to a 20VDC supply (multimeter reads about 20.5V), no load, I can turn the voltage control pot (labeled TR2, above) about half way up, and all looks good as the voltage climbs to about 12.7V, then the voltage stop climbing as I turn the pot the rest of the way and stays constant. The pot controlling the current seems to work fine (tested with a 10 ohm resistive load). Any adjustment to the current pot has no effect on the output voltage (which is as it should be, I think).

The voltage should climb to Vin (~20V) minus about 2-2.5V. When I measure the voltage on pin 4, Vref, it is about 2.77 (almost perfect from the spec sheet) up until the pot hits that 12.7 Vout mark, then Vref starts decreasing and Vout stays constant as I further turn the pot.

All three strings exhibit identical behavior.

The way I understand this chip to work is that it uses Vref as a (what else?) reference voltage, and holds it constant, causing Vout to rise as TR2, and therefore the ratio TR2/R4, increases. This is not happening, or is only happening to a point. Can anyone see why?

I'm going to see if I can further parameterize the problem this afternoon, figure out exactly where TR2 resistance is at this stop-point and measure anything else that seems relevant, but thought I'd get this out there in case it's already been seen. Also, I can accommodate any response that has additional measurement points.

TIA,
Allen
 

crutschow

Joined Mar 14, 2008
34,284
Note that the LM741 is a poor choice in this circuit since its inputs cannot work down to 0V. Check its output to pin 2 of the L200 while varying the output voltage to see if it's doing anything odd. The voltage on pin 2 must be within <0.45V of the output voltage for the L200 to not limit the output voltage.

To insure proper operation over the full output voltage range you should use an input and output rail-rail op amp.

Otherwise I would double check all the wiring connections compared to the schematic and that all component values correspond to the values in the lower component list.
 

Thread Starter

anw

Joined Jan 11, 2014
10
crutschow: Thanks! Great idea. I didn't check that, and I will, and will respond. I may have to breadboard the circuit (again!) to get to all the pins, but I'm willing!

MrChips: Thank you for helping me out there. Not to go OT t0o badly, but can you tell me how you did that? I've been a denizen of lot of fora, but until now they've always been about programming or sysadmin. I can see that's gonna change!

Thanks again to you both...

Allen
P. S. MrChips- I like your sig. I've found that to often be true in my case.
 

MrChips

Joined Oct 2, 2009
30,712
Here is how I get a picture to appear directly in the post:

Step 1:

Go to the website showing the image. Right-click on the image.
Select "Copy image URL".

Step 2:

In your post, click on the "Insert image" icon at the top of the message box.
Paste (Ctrl-V) the URL into the dialog box. Click OK.


(There might be some restrictions for a New Member until you reach 10 posts.)
 

Thread Starter

anw

Joined Jan 11, 2014
10
Thanks again. I wanted to put a screen shot of my KiCad schematic in the post, which showed all three channels, but I don't see a way to do that.

Off to measure voltages, resistances, etc.
 

Thread Starter

anw

Joined Jan 11, 2014
10
OK, here are the results of my test. I'm attempting to paste an Excel spreadsheet table in here...

Didn't work. Try as a pdf...nope. I've attached the pdf table (I think). I wanted to put it inline so I could refer to it, but oh, well. I guess I need to go back to Forums 101.

So, attached are the measurements at various points, pot levels, etc.



Two things come to mind:
1. I don't understand "the inputs will not work down to 0 volts". The only place there may be 0 volts is across pins 2 & 3, right? And that's what an op amp does, is amplify what's across the differential inputs, in this case pins 2 & 3, right? So, what would be wrong with 0 volts across pins 2 & 3?

But, be that as it may, the measurements are in the attachment, and there's nothing that's 0. The 2/3 differential goes down to 60 mv, but when it's that low, it's in the range the circuit is working.

2. This circuit came from the L200C data sheet, but the 741 op amp specs say absolute max input voltage is 18 V. I want this to work the full range of the L200C, up to 30V input, so I put a 7812 in to power the op amps with 12 V (plus to run a fan). You'll see the measured supply to the op amp: 11.85 or something like that, so its supply voltage never goes to the L200C max (though it does go up to over 12 V, which I'm not sure I understand, since it's only operating on 11.85V; next step is to turn this into a perpetual motion machine).

However, this is not the part of the circuit controlling the voltage, and the current part works fine (at least to the extent I've tested it; don't think I've attempted to run it all the way up to 2 amps- it gets HOT!).

There are two documents: the L200C datasheet, where this circuit is found, and a "Design Guide". I'm not sure why the op amp is even in there, and none of the current limiting circuits in the design guide have an op amp as an input to pin 2. As you can tell, I just copied it (assuming expertise on the part of others), but I wonder why you can't just use a pot on pin 2, especially now that I've seen the voltage range there.

More thoughts?

TIA,
Allen
 

Attachments

Last edited by a moderator:

MrChips

Joined Oct 2, 2009
30,712
If I need to create my own graphics, I take a screen shot by pressing "Print Scr" on my keyboard.

Then I use Corel Photo House to edit and save the image.
In Corel Photo House, I go to the File menu and select "New from Clipboard".

After I have edited, cropped and saved the image, I upload the image to an album in AAC. From there I can embed the image in a forum or blog. Some restrictions may apply for a new member.
 

MrChips

Joined Oct 2, 2009
30,712
I don't know what restrictions apply to a New Member (posts fewer than 10).

Scroll to the top of your AAC screen. On one of the menu bar you should see User CP which stands for User Control Panel.

In your Control Panel you should be able to see what you can do, for example, I have Pictures & Albums.
 

Thread Starter

anw

Joined Jan 11, 2014
10
Wow! That's great, the way you cut in the table. I went to User CP, and didn't have a "Pictures & Albums", so maybe when I hit 10 posts...I did reply to someone else's just now, for what it was worth. He was trying to do something I also tried and never could make work. Anyway, one more post toward the limit...and thanks again!
 

crutschow

Joined Mar 14, 2008
34,284
So, attached are the measurements at various points, pot levels, etc.



Two things come to mind:
1. I don't understand "the inputs will not work down to 0 volts". The only place there may be 0 volts is across pins 2 & 3, right? And that's what an op amp does, is amplify what's across the differential inputs, in this case pins 2 & 3, right? So, what would be wrong with 0 volts across pins 2 & 3?
I'm taking about the input common-mode voltage to ground, not differential voltage across the pins.

If you look at the pin 2 voltage at the maximum output voltage you will see that it is about 0.46V below the supply voltage, which is the point where the current limit kicks in, so that is what's limiting the output voltage. For whatever reason the op amp is lowering the pin 2 voltage to the point that the current limit is operating. A rail-rail op amp shouldn't do that.
2. This circuit came from the L200C data sheet, but the 741 op amp specs say absolute max input voltage is 18 V. I want this to work the full range of the L200C, up to 30V input, so I put a 7812 in to power the op amps with 12 V (plus to run a fan). You'll see the measured supply to the op amp: 11.85 or something like that, so its supply voltage never goes to the L200C max (though it does go up to over 12 V, which I'm not sure I understand, since it's only operating on 11.85V; next step is to turn this into a perpetual motion machine).
The 741 will operate up to 36V maximum with a single supply (±18V dual supplies) and the input can operate over ±15V or 0-30V maximum with a single supply.
There are two documents: the L200C datasheet, where this circuit is found, and a "Design Guide". I'm not sure why the op amp is even in there, and none of the current limiting circuits in the design guide have an op amp as an input to pin 2. As you can tell, I just copied it (assuming expertise on the part of others), but I wonder why you can't just use a pot on pin 2, especially now that I've seen the voltage range there.
The op amp was added to reduce the voltage drop across the shunt sense resistor R3 and to allow the use of a high value pot to adjust the current limit. If you don't mind a .45V drop across the resistor at the current limit point and using a low resistance rheostat for R3 to adjust the limit then you can do away with the op amp.
 

Alec_t

Joined Sep 17, 2013
14,280
If you use 30V, rather than 12V, as the supply voltage for the opamp you will be able to increase the output voltage.
 

djsfantasi

Joined Apr 11, 2010
9,156
You don't have to create an album to add pictures or diagrams to your posts. You can attach them and then embed the attachment in the body of the post. I describe the process in my blog.
 

Thread Starter

anw

Joined Jan 11, 2014
10
Ha! Thanks, djsfantasi, I will make good use of this new-found super-power, starting right now:



Doh! I've looked at that datasheet a thousand times, and never noticed the "±" on the max supply. I had every intention of using my new-found super-power and pasting in this screen shot, and saying "What do you mean, 36V?". Ah, the poetic justice that accrues to the arrogant! And, fixing this is another reminder of why God gave us Dremels & jumper wires!

So, crutschow & Alec_t, armed with Dremel & copper, I will go give that a shot right now; well, after demonstrating further ignorance for a bit, though now I hope it's purely academic:

I'm still not too sure what you mean by common mode voltage. I've always associated that term with an interference (be it AC or a DC bias) that shows up across two or more differential pins or wires. I'm not sure I see where that's happening in this case. On the other hand, maybe it's because I don't understand this:
If you look at the pin 2 voltage at the maximum output voltage you will see that it is about 0.46V below the supply voltage, which is the point where the current limit kicks in, so that is what's limiting the output voltage. For whatever reason the op amp is lowering the pin 2 voltage to the point that the current limit is operating. A rail-rail op amp shouldn't do that.
The pin 2 voltage is 12.32V, the supply voltage is 11.85V, so the pin 2 voltage is .47V above the supply voltage (assuming you mean supply to the 741 and not the 19.95V to the overall circuit). Is this a typo? And where does common mode voltage apply?

And, by the way, I still don't understand how the op amp is putting out more voltage than its supply; given the numerous occasions that I've been thwarted by the 1st Law of Thermodynamics, I'm mildly interested in that.

I will point out that I'm fully willing to accept the root cause of these questions is that I'm just dense, as long as it comes with an explanation.

And, speaking of dense, or, perhaps more accurately, ignorant, what is a "rail-rail" op amp? I've never heard of such a thing. I've heard the colloquialism "rails" applied to power & ground delivery, often use it that way, but not op amps.

Though I'd like think these answers and my immediate testing of the op amp voltage theory will close out this thread, I've been known to be wildly over-optimistic. And, in another fit of wild optimism, I promise I'll come back and let y'all know that now my circuit is functioning perfectly.

And to all, my most sincere appreciation for not only the technical expertise but the forbearance with and rectification of my limited forum skills, which are now improving daily.

Highest regards,
Allen
 

Attachments

crutschow

Joined Mar 14, 2008
34,284
..............................
I'm still not too sure what you mean by common mode voltage. I've always associated that term with an interference (be it AC or a DC bias) that shows up across two or more differential pins or wires...............
Common mode means the voltage is common to both terminals. It's not differential (across) the two terminals.
The pin 2 voltage is 12.32V, the supply voltage is 11.85V, so the pin 2 voltage is .47V above the supply voltage (assuming you mean supply to the 741 and not the 19.95V to the overall circuit). Is this a typo? And where does common mode voltage apply?

And, by the way, I still don't understand how the op amp is putting out more voltage than its supply; given the numerous occasions that I've been thwarted by the 1st Law of Thermodynamics, I'm mildly interested in that.
The common mode voltage is the voltage is common to both the 741's inputs.

I'm was referring to the output supply voltage not the 741 supply.

The output voltage is pulled up by R1.
And, speaking of dense, or, perhaps more accurately, ignorant, what is a "rail-rail" op amp? I've never heard of such a thing. I've heard the colloquialism "rails" applied to power & ground delivery, often use it that way, but not op amps.
It's an op amp that can operate with the input common-mode voltage and/or the output voltage going to either of the rail voltages. A 741 can do neither.
 

Thread Starter

anw

Joined Jan 11, 2014
10
Well, thanks again, and I'll let you know how this goes, hopefully tonight. Don't by any chance have a part number for them thar rail-rail op amps, do you?
 
Top