Adjustable Bench Supply won't adjust (at least not as much as I'd like...)

[CurtZHP]

Recently built an adjustable bipolar power supply with a separate 48VDC rail to keep on the bench for tinkering with audio circuits.  (Schematic attached.)

This is based off of a fixed-voltage bipolar supply I've used successfully numerous times.  All I really changed was the resistor network around each of the bipolar regulators (LM317 and LM337).  I followed the datasheets and put a 120 Ohm resistor between the regulator's adjustment pin and output, with a 5K trimmer between that and ground.

Here's the problem....
The positive rail works fine.  I can adjust its output anywhere from a few volts on up to about 35VDC.
The negative rail is the issue.  I can adjust down to a couple volts, but the absolute highest it will go is about 11VDC.  Input voltage is about +/- 37VDC.  At first, I was blaming the trimmer, because that was a salvaged part from who-knows-where.  Most of the other parts in this supply were purchased new for this project, including the LM337.  I replaced the trimmer with a brand new one, and the result is the same.

I even tried putting a load on the supply to see if that mattered.  It didn't.  I also tried, just for grins, adjusting the positive trimmer while measuring the negative rail, just to see if there was some odd interaction going on.  Didn't make any difference, which I suppose is a good thing.

Checked for shorts or any other such boo-boos.

At a bit of a loss.  It's probably something stupidly simple.

Help?

 

[abbey road d enfer]

Beware that LM317 and 337 have a different pin-out. Vin and Vout are reversed.

 

[CurtZHP]

Beware that LM317 and 337 have a different pin-out. Vin and Vout are reversed.

Took that into account.

Found the problem, by the way.

Although the schematic is correct with regard to C117 (the noise filter cap for the negative regulator), I somehow managed to reverse it in the actual build.  Tiny cap, so it was hard to spot the mistake.  I guess I never had it plugged in long enough for the darn thing to explode in my face.  Flipped it around, and now the negative regulator behaves as expected.

I knew it'd be something simple!