TREAD outputs 1/2 expected voltage

How about TP1 and TP3? TP1 and TP2 only measures the voltage difference before and after the LM317. TP1 and TP3 is the DC voltage after rectification.

Measuring DC on an AC source is tricky.

First, most cheap meters can't do it right at all.  Their DC function can't filter AC and give you only the DC component.  The higher the ripple component, the greater the error in their reading.

Second, you don't talk about what caps are connected.  If your report is complete, you are getting a large DC measurement purely from rectified AC, which is wrong: it's still AC, just not sinusoidal any more.  You only get DC when you put a big filter cap on the bridge's output, to act as a power reservoir.  If you have C5 installed in the TREAD, then the first explanation may hold, and explains the difference.

If you had caps on the bridge in both cases and you're using a meter that really can extract trustable DC measurements from a signal with a lot of variation, then I'm at a loss.

Quote:

Originally Posted by zenon 

These are my test results:

P1 - P2

DC: 11.5

AC: 0

That says you're making the regulator drop 11.5 VDC. That's pretty wasteful, but it may be okay, depending on the current you're drawing. You did run your config through the estimator, right?

A small AC component is expected here. You should be measuring on the mV scale to get a nonzero value.

R2 looks like 1.5K?  I can't see the R1 markings.

Does twiddling VSET change anything?

By the way, it's "TP1" and "TP2", as in "test point".

Your meter is lying to you, or you aren't measuring what you think you are.  The DC component is sane if R2 is 1.5K and R1 100 or 120 ohms. The AC component simply cannot be that high if the DC value is what you say, unless it's going through ground, and that can't happen in a circuit like this unless you're doing something very weird.

If you're getting 42 mV AC here, that's a problem.

What do you get for VDC between TP1 and TP3?

Are just you concerned by the measurements, or have you hooked it up to a load and observed some other problem?

The most common thing to get backwards are the tantalum caps.  Their positive lead is the marked one, not the negative, as with electrolytics.  Your pics don't show the markings, so I can't say whether they're in right or not.

Those supply wires...are they connected to IN+ and IN- on the bottom side of the board?  I don't see any solder peeking up to the top side.

Is there an insulator and shoulder washer isolating the regulator from the heatsink? Although this will not cause a problem with circuit operation it is considered good practice to isolate the two.

Quote:

Originally Posted by zenon 

These are my test results:

P1 - P2

DC: 11.5

AC: 0

 

P2 - P3

DC: 19.25

AC: 41.7

 

I reviewed the schematic and realize that P1-P2 doesn't measure what I initially thought it did.  But the P2-P3 AC component is concerning.  Ideas?

 

Attached are some photos, I don't think there's anything installed backwards.