Here is an easy MOD I applied to my Talking House II for better low end frequency response.
Here is an easy MOD I applied to my Talking House II for better low end frequency response.
Capacitor C301, .047 uf, couples audio to the modulator op-amp. It is in series with a 4.7 K resistor which gives a lower cutoff frequency of around 750 Hz.
The "Direct" audio input is applied to C301. The modulator op-amp is direct coupled to a follower transistor which modulates the final RF output stage. As such, the RC network (C301 and the 4.7 K resistor) is primarily what defines the lower cutoff frequency.
I replaced C301 with a 1 uf capacitor. That should lower the frequency response to around 40 Hz. C301 is located nearly in the middle of the PC board. I modified two of my TH transmitters as stated. In both cases C301 was green in color and "Chicklet" sized.
The result; a much better low end audio response. No low end has always been the complaint for early Talking House models.
If you missed the post of the TH schematic it's HERE. The parts values shown are not what my TH had but the schematic layout seems correct.
Page down to figure 4A. Look for an op-amp with the location numbers 90, 94 and 95A pointing to it. Follow the inverting input to the resistor and capacitor in question. Or look at the "Direct" input and follow the wire to the right to the capacitor.
I used a .1uf ceramic disc cap. Works perfectly and doesn't load down the op amp.
RFB
I dunno why this is a problem ... I'm loathe to make any changes to the circuitry because I can't find anything wrong with the low frequency reproduction on my current in-use-daily TH5. Could it have something to do with part tolerances between mfr runs? I think it might, but I have no real evidence. All I know is it's hard to find anything wrong with the audio as I receive it in the taxi I drive, 1997 Chev Astro.
I think I posted it before:
http://fhtinyradio.com/Library/AudioLib/FHTRTaxiAstrovan082211.mp3
...recorded inside the van on a Tascam DP-008 built-in mics, situated approx. shoulder level between the front seats ... about 3/4 mile or so away, I think (with town buildings and traffic in between), late afternoon.
"Could it have something to do with part tolerances between mfr runs?"
Its possible. There have been a few runs of revised units since the first introduction of the TH V5.
But in regards to the audio mod for increased low end response on the V5, C301 is NOT a frequency cut off component. It is merely a coupling capacitor between the direct/internal audio and the op amp IC input. Components C337 and R302 set up the op amp's gain and frequency response.
I'm surprised no one has caught onto that nor admitted to when replacing C301 with an electrolytic 1uf cap, it loads down the op amp so much it creeps up with distorted audio for the first several minutes. Not exactly a correct modification to begin with since it is not the component setting up the op amp's gain or frequency response.
But your right the V5 unit does have decent audio on its own with the .047uf cap in C301's slot. Replacing that with the .1uf ceramic disc smooths out the existing response a bit in the 400hz-6khz range. As far as the low end, its fine the way it is.
But for those who want to try to increase low end response, I suggest examining the datasheet for the op amp IC and follow the test circuit examples with particular attention to the feedback loop component...a 10M ohm resistor and a .1uf cap at the input. A good starting point and take note of it and the circuit in the TH and see where they set up the cut off with C337 and R302.
UPDATE: Just for kicks and giggles I decided to open up one of my TH units which has the .1uf ceramic cap in place of the 047uf stock cap at C301. Put the 047uf back in, and removed C337. Guess what. The unit has a full frequency response from 10 cycles out to 22khz.
It almost follows the curve in the test circuit on the datasheet, given R302 is not a 10meg resistor, but if it was that op amp would follow the test curve exactly.
For standard AM bandwidth a frequency range between 50hz and 8 or 10 khz is optimum. Just throwing a wild guess out there, I think a .020 or so for C337 would put the unit in that ball park.
For the nth decimal place answers..or rather steps to the correct answer...here are a few articles on op amps and setting up frequency response and gain.
Lecture 29 Op Amp Frequency Response
Google "setting op amp frequency response", without the quotes and there is a ton of valuable info.
RFB
TH ver5 was supposedly redesigned to address the audio response. As such it probably does not require modification. If it ain't broke, don't fix it.
As for coupling capacitors and op-amps:
A closed loop inverting op-amp forms a virtual ground at the inverting input. As such, the input impedance is a function of the input resistor and coupling capacitor. Essentially, the virtual ground puts the RC network between the audio source and ground.
That input RC network does indeed form a high pass filter whose cutoff frequency is equal to .159/RC. As with any "coupling" capacitor, if too small a value is used, low frequencies will be attenuated as frequency goes down and Xc goes up. That's why we "recap" amplifiers as they age.
If the op-amp input bias current were unusually large, the capacitor would cause an output offset voltage. However, with the proper value feedback resistor this is generally not a problem.
The feedback resistor between the output and inverting input sets the closed loop gain of the op-amp stage as a funtion of the ratio of Rf/Rin. When a capacitor is placed parallel to the feedback resistor it forms a low pass filter. As frequency goes up, Xc goes down, reducing the gain of the stage. If the capacitor is removed as suggested, yes the high end response will go up and exceed the allowed audio bandpass for an AM station. At 22 kHz, the bandwidth of your transmitted signal will be 44 kHz and splatter adjacent channels.
There are some special case op-amps such as the Norton op-amp whose input stage is primarily current controlled but in closed loop operation even those have similar characteristics when properly designed.
I taught Linear IC devices to include op-amps at the National Institute of Technology for 7 years. The information presented is from their certified curriculum.
As for the C301 mod, I have experienced no distortion. The low frequency response has indeed improved.
Increasing the feedback capacitor would reduce the high frequency response giving the illusion of improved low end response as you increase the amplitude of the audio input to compensate.
Don't believe me READ HIGH PASS FILTER HERE and LOW PASS FILTER HERE and see what they say about op-amps with RC networks.
Hmm.. must be different laws of physics going on cuz none of those mods put out there did anything to my TH unit except bog down the op amp.
Now modifying the feedback loop with a different resistor and capacitor set the op amp where it should be..and the correct way it should be done. The stock C301 or a 1uf or .1 had no change in that op amp's frequency curve and gain.
I wont argue about different methods of setting up op amps, to which there ARE several different ways, including the references I posted as well as yours. So byte me! 😛
Besides..when an op amp is set up for proper gain along the frequency response curve via the feedback loop, there is no need for external filtering, hi pass or lo pass. According to my sweep generator and audio spectrum analyzer, the mod in my TH is correct and no where near the distortion of the stock or the other wacky mod method.
All I know is what I did actually worked the right way for the circuit the way its designed. If another method works for you or for others...that's great! As long as mine works and yours works and everyone else's works...it's a sure byte outta the low end. 😛
PS...the TH units I have are not the modified versions for improved sound...in case that may have been in question.
RFB
Thanks Guys!
I was about to send this thing back and then I stumbled across this thread. I did the mod and now i have beautiful treble where before it was so muffled I couldn't stand to listen to it. Just had to create an account to say thanks so much!!
KC
"Just had to create an account to say thanks so much!!"
Your very welcome and welcome to the forum! I hope that you will make use of the account other than to just say thanks. 🙂
RFB
My Talking house (which is supposed to be the improved sound version) had a 10 mfd electrolytic in C301. Here was my mod I put a .047 at C301 and left C337 in. Without C337 being in (which was in my board a 47 pf) my TH would oscillate and squeal like a pig so C337 has to stay. Now I have plenty of treble and good bass response too. Good enough to go to broadcast music now.
I was scratching my head for a while after cutting C337 out of the circuit but this cap is necessary for proper operation of the audio section. Now I'm perfectly happy with this transmitter. They have it right with the output but their audio mods were just way the heck off. Who's tone deaf ear approved this 10 mfd cap in C301????
KC
Thanks RFB for the warm welcome. I think I'm going to like it here.
KC
i'd like the mods to turn a iAM ver1 into an iAM Ver2.
I have a iAM ver1 it can only mod 60-80% and has AGC. basically it's a THII with the audio enhancement.
the iAM ver2 has AGC disabled and can modulate to 130 percent in addition to the audio enhancement.
this is what i have been told by a few people.
my concern is what needs to be done in order to get that 80% up to 130% and i have already been told how to bypass AGC