Here is a PCB layout. I made it single-sided and somewhat spacious for those who want to etch their own PCB. The input will accommodate a two-screw terminal block (Onshore Technology part# OSTTC022162) or a coaxial power jack. The output is laid out for a two-screw terminal block. Of course, you can solder wire leads directly and forego the connectors altogether. The pot is a top adjust trimmer (Bourns part# 3362P-1-502LF). The mounting holes are .125" for 4-40 screws.

Let me know if you see any errors or have any questions.

 

BTW: I was fiddling with mine today to check the set voltage and accidentally tried to reconnect the battery backwards and there was no damage so it does seem to be idiot proof.....

The set point voltage should be 1.40 V/cell in the battery.

 

The 9V battery has a polarity connector so it only goes on one way. I suppose somebody could try to force it on reverse, never worried about that.
...

Actually I think we were both referring to the battery symbol being drawn with reverse polarity on your schematic.

...
I suppose you could add a 1N4148 blocking diode between the right end of R1 and the top of R2 to replace that trace? That sound right?
...

Diodes won't protect a charger from a reverse polarity battery. The fault current is still in the same direction as charge current, so the diode will conduct anyway.

If I could make one suggestion, why bother with the 7815 regulator? This forces the need to use an unusually high DC voltage (ie 18v) and does not seem to gain much. Since the charge current is only about 10mA why not just use a series resistor from the plugpack?

If you are worried about regulation, the plugpack voltage would not sag much from a 10mA max load. So regulation is not really needed? The only down side is that the resistor would need to be chosen to suit whatever the DC plugpack voltage is. But if a standard 12v DC plugpack puts out 14v etc with a 10mA load then the resistor calc is very easy and only needs to be done once.

 

If I could make one suggestion, why bother with the 7815 regulator? This forces the need to use an unusually high DC voltage (ie 18v) and does not seem to gain much.

Without it, the current through the shunt reg is variable and it affects voltage accuracy. It would "work" without it, but I wanted better accuracy. Dealing with such small current, the power loss for the 7815 is miniscule.

If you are worried about regulation, the plugpack voltage would not sag much from a 10mA max load. So regulation is not really needed?

But it wandered a lot with temperature. I recall I put in the 7815 to stop the drift.

 

I found the hand drawn schematic of the final design after I added some additional LED indicators to make it easy to see when the battery was almost fully charged (see attached). The amber LED lights when the battery is about 80 - 85% fully charged, the yellow one lights at about 90 - 95%. Doesn't change operation of the basic circuit, just indicates state of charge. The brightness of the red LED still indicates gross state of charge, the other two signal near the end.

 

I was actually planning on designing a 9V battery charger pcb.
the battery is actually the Powerex 9.6v 230mah battery will this battery be ok with this circuit?
Tracecom it looks like you are using Diptrace judging from your schematic and board layout. I too use diptrace. would it be possible to post the diptrace files.

Thanks!

 

I was actually planning on designing a 9V battery charger pcb.
the battery is actually the Powerex 9.6v 230mah battery will this battery be ok with this circuit?

I think you will need to increase value of R3.

EDIT TO ADD: the Powerex ad calls it a "true 9.6V" battery which may be an eight cell Ni-MH which would require a charging voltage of

8 X 1.4 = 11.2V

If so, increase R3 to about 13K