[Ron Ramirez]

...probably should ask the moderator to move this part into the proper thread

Done.

[rfeenstra]

Don't forget, the early radios used a semiconductor detector, as morzh said!

[Jake Blake]

WHAAAAAAAAAAAAAT!!!!!!!!!!!!!!?????????

Satan, get thee behind me!!!  

[Jake Blake]

Oh. . .  . the crystal sets.............   ok

[mikethedruid]

Here's one of several "solid state" receivers in my collection...


[Image: https://66.media.tumblr.com/2f49154f3388...1_1280.jpg]

[Jake Blake]

Waaaaaaaaaay cool!    Ok.  Thanks for sharing.  

[rfeenstra]

Installed 35's in the appropriate 3 locations (instead of the 24A's).  The radio works great.  distortion is gone and very strong stations ( like my local transmitter) do not cause distortion like they did with the 24's in the circuit.  So far, very pleased.  Certainly not very original, but at least I can listen to it for hours now.

[rfeenstra]

In a post in Ron's 70 modifications, I mentioned that I compared the sound of the 570 with my 96 TRF radio.  The TRF has much better high frequency response as the there are no highly selective IF transformers to kill the side bands.  As a result, cymbals and snares sound a bit dull on the 570.  I decided to play with tuning a bit to see if i could broaden out the IF response to improve the high frequency response.  I used only the three compensating caps to play with that.  At the expense of sensitivity and selectivity, I was able to improve the high frequency response enough to actually get a little shimmer on cymbals and some snare sound rather than just a thud.  Here are some shots of the band pass characteristics taken from the output of the second IF transformer at the detector diode.

First, the curve with the radio properly tuned.  The side markers are at +- 5 khz.  Notice that a 5khz signal is attenuated by about 7 db. 

   


The second photo shows the curve with the IF compensators slightly detuned plus and minus.  Notice that peak signal level is down by about 6 db.

   


In this photo, the markers are at +- 8 khz and the IF compensators detuned a bit more.  Peak signal level is down about 10 db.

   


Then for fun, the curve with an RF signal modulated at close to 100% with a 8 khz sign wave.

   

[Ron Ramirez]

Interesting. Of course high fidelity wasn't even on the radar yet in 1931, so no surprise the IF transformers have limited bandwidth when peaked according to the manufacturer's instructions.

This would change a couple years later.

[rfeenstra]

Interestingly, I set my 70 (unmodified) next to the 570 and fed it the same signal.  Aside from the typical 70 issues, it has a better high frequency response. So, now I'm going to do some audio testing on the 570 to see if the audio amp section is rolling off the highs.  Or maybe my Celestion speaker cone is not so good!  I may have to swap speakers to see what that does.  Also, maybe I'll change that 220 pf cap to a 100 pf in the detector circuit.  Lots of stuff to play with.... fun!