[SteveJB]

I just took out the reciever to get ready to reacpp it. There is a box on the reciever chassis on the right front side next to the tuning cpaacitor. Does anyone know what this is? It looks like a filter block that I have seen on the older Philcos, but I am not sure as it is not labeled on the documentation I have. I hope to get a better look tomorrow.

Also, for the tweeter wiring error, do you have to use non polarized caps? Can I use an electrolytic? Why use a 4.7 mf or 10 mf instead of the 2.2mf?

When changing the tweeter wiring, the schematic shows the tweeters physically close together. Of course they are on opposite sides of the case in the radio. Where do you actually make the connections for the white and green wires? Under the power chassis where the wiring from the tweeter leads to?

Thanks for your help,
Steve

[Ron Ramirez]

Hi

The "box" you refer to is probably part (108), which contains three capacitors - two 0.1 uF and one 0.7 uF. These caps are part of the AFC (magnetic tuning) circuit, and are critical. You can get close to 0.7 uF using new parts by connecting a 0.47 and a 0.22 uF in parallel to achieve 0.69 uF, close enough.

Regarding the tweeter wiring error, refer to the page on this subject on my website.

http://www.philcoradio.com/tech/tweeters.htm

It is preferred that you use a nonpolarized electrolytic for this speaker crossover cap. Some guys advocate using two regular electrolytics connected in series, with the positive leads tied together and the two negative leads being used as the leads of the original cap was used. But I would prefer a nonpolarized electrolytic, which is used for speaker crossovers.

Why a 4.7 uF or 10 uF? As stated in the text on the page linked above, the original circuit of the 38-690 used a 1 uF cap, which yielded a crossover point of 19000 cycles (Hz). Way too high for this radio. A 2.2 uF cap will yield a crossover point of 9000 cycles; 4.7 uF, 4000 cycles; 10 uF, around 2000 cycles. You see, the higher the value of the nonpolarized electrolytic used, the lower the crossover frequency.

Crossover frequency is defined by the McGraw-Hill Science & Technology Dictionary as the frequency at which a dividing network delivers equal power to the upper and lower frequency channels when both are terminated in specified loads.

The 38-690 tweeters should be receiving the audio frequencies above 2000 to 4000 cycles, not 19000 cycles. At 19000 cycles you will hear very little from the tweeters. At a 4000 cycle crossover, the tweeters come to life in a 38-690. At 2000 cycles, they make their presence known even more.

I would not go below 2000 cycles for a crossover frequency. In fact, I think I would stick with a 4000 cycle crossover. The woofer in this set should be able to reproduce the low end audio up to 4000 cycles.

The wiring change for the tweeters, if it needs to be done, is done under the chassis of the audio amplifier/power supply.

[Ron Ramirez]

One more thing. There is a bakelite block part in the tuner chassis, part (97). That is not a capacitor. That is the 10 KC filter coil. DO NOT DISTURB IT. LEAVE IT ALONE.

By leaving it alone, I mean don't pull it out and attempt to remove its insides. You will have destroyed your set's 10 KC filter coil if you do so.

So remember...this is very important...do not disturb part (97).

[SteveJB]

Thank you Ron. I am taking my time on this one to be sure I get it right.
Steve

[SteveJB]

I just finished the electrical restoration and used the 4.7 mf caps in the tweeter wiring. The radio plays very well. The only problem left is to get an original style grill cloth and one of the tweeters is blown. I did try to recone it, but I need to try again, the voice coild doesn't move in/out welll and doesnt sound so good as the other tweeter.

Do you know where I can get an original style grill cloth for this model?

Thanks,

Steve

[Ron Ramirez]

Yes - it takes pattern number 14 from Radio Grille Cloth Headquarters.

http://www.grillecloth.com/

The "ribs" (stripes, if you prefer) in the cloth run vertically, not horizontally.