diyAudio (
-   Power Supplies (
-   -   TROUBLE WITH THE L165! (

teleman 10th July 2013 01:52 AM

1 Attachment(s)
I have now spent nearly 4 days trying figure out a persisting problem with this L165 virtual earth circuit! No matter what i did, both the +/- voltages are quite assymetrical giving +11.84 & -12.08 or thereabout.

I even removed R3 & R4 as it was on an other variation of this circuit & it got worse! This particular reg. is a ST brand (Micro Electronics?). I also read elsewhere of d.i.yers experiencing the same problem with this particular device. I even built 3 seperate circuits, still the same problem persists on all of them. Could it be that this particular brand or batch is faulty?

Do any of you guys have any idea or suggestion as to what this problem might be due to or suggest a similar high current feedback amplifier(s), which can source at least 100mA?

teleman 10th July 2013 04:53 AM

1 Attachment(s)
Ahh....just remembered something! Since none of the scheamtics show the dc by-pass caps, I too didn't include these in my circuit! Perhaps this is what triggering oscillation thereby causing this problem???

Can anyone spot a fatal mistake in the schmatic on #post 1 ? No prizes given though!

Mark Johnson 10th July 2013 03:23 PM

1 Attachment(s)
The correct value of R3 is Infinity, the correct value of R4 is Zero. Post #1 in this thread incorrectly sets them to (R3=1K, R4=22K).

Have a look at this figure which I copied from the IC datasheet:

teleman 10th July 2013 04:23 PM

Thanks a lot! I did remove R3 & R4 & just linked these as in fig.2 but It just got worse, with the values shown in Fig.1! I will change these as shown & report back. perhaps these new values might help! It's just can't seem to get it exact on both rails!


Mooly 10th July 2013 05:26 PM

How in-exact is it ?

I notice the data sheet says -/+6 volts as a minimum and then shows an application using -/+4 volts. It also in two diagrams has the inputs mixed up. Look at your first post and post #3. The + and - inputs are reversed. Post #4 is electrically drawn correctly but I don't know if the pin numbers are correct

The unity gain follower uses a 1k resistor between the inputs (which is technique to reduce the available gain) and this points to stability issues. You could alter the circuit to do the same with two resistors and then add a small (say 47pf) cap across the "feedback" to kill any tendency to oscillate.

BZed 10th July 2013 05:31 PM

Why not add a small pot in series with your reference between R1 and R2 of 100 to 500 ohms and connect the wiper to pin 1. Bypass it with a cap for noise reduction then adjust for the output you want. The output voltage is going to be determined by how well matched the reference resistors are. If they are not perfect there will be an offset. This way you can adjust the offset out.

teleman 11th July 2013 12:46 AM

Transistormarkj, the circuit on #3 doesn't work & blowing fuses! I actually built this twice but with the same results both times! There are in fact 3-4 variations of this on the net! Only the #1 circuit works with the + rail 11.84v & the - at 12.08v from 24v supplied from 4 x 18640 batt. & through a buck booster. Tried with the dc caps too still no improvement. The voltages are constant but this annoying assymmetry! The resistors are matched of course! Yes, I did notice the wrong pin orientation & corrected it. I actually got the specs from ST Micro Electronic pages.

There's definitely something untowards going on either with this chip or the various circuits!

ST Micro

Mooly 11th July 2013 06:23 AM

1 Attachment(s)
Take your post #1 circuit and try this.

teleman 11th July 2013 07:36 AM

A vast improvement! :) with the reading +11.84 & -11.91!! Thank you...thank helped a lot! but does this still mean high dc offset?

Mooly 11th July 2013 08:09 AM

Unequal supplies have no effect on (for example) opamp circuitry. If you have +16 and -6 or -18 and +5 the DC offset of the circuitry is essentially unchanged.

The fact that the results have improved/altered possibly points to the original circuit oscillating. A scope check would really be needed to confirm that but the modified circuit should prevent any possibility of that happening. The values aren't critical. Also the 1 ohm and 0.22uf isn't strictly necessary or lets say if you want it then I would also add the same to the other rail too. Maybe increase the 1 ohm too.

All times are GMT. The time now is 07:10 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.
Copyright ©1999-2017 diyAudio