CONTINUED!
From CB Tricks - 2016 DTD...
Quote from: Reverend Bow on October 20, 2016, 12:13:32 PM
Hey Andy,
Thanks for the reply.
Don't worry about my SS3900's Audio Bandpass, I have that covered... ;D
The Graph below is a sweep (with a Function Generator) from 10Hz to 10kHz of the transmit circuit on the SuperStar 3900/Cobra 148 GTL/EX The High end caps (bypass to ground) are stock, so the biggest limiting factor to stop the Audio bleed into the next channel up or down is the Mic Element. The Red Line is the the Superstar, the Yellow Line is the Icom IC-718 in AM Mode.
Would like to know the impact the Electret and it's frequency response would be in comparison to.
Plus, would need to know how the Icom's own support (Modulation Method or Theory of operation would be nice for those playing along at home) circuitry works - at least we'd have a comparison. At least to be fair.
In the SS3900 - are you sure you want to use this platform? I see considerable rolloff past the 3KHz and above, not much is needed after you start to go above Tenor and Alto voice ranges - unless you're looking into broadcast quality. Then there is an issue here with the RF output and neutralization that is needed. (Heavy Capacitance)
So, before we start to compare Apples to Oranges - let's make sure they're both "fruit" ;D
If there and Is(?) a way, to post these results I'd love to review what my efforts show - I can't do much here at the moment because I'm locked into being a caregiver to an elderly patient...so I cannot - at this moment in time - reproduce or verify, any of your results.
All I have at this juncture is the memories and "direct side-to-side" checks before and after conversion on any given CB radio "PC-68", PC-122 Cobra 29 (4-Pin) Realistic TRC-453 and TRC-474 (5-pin DIN) Cobra 148 and Uniden Grant XL (5-pin Amphenol) - to verify keying and audio loudness effect, if that matters. Connector reference can be found here...
The thread was developed for corrections, additional notes, factory revisions and changes to the mic -amp circuit throughout the course of it's life.
Regards!
:+> Andy <+:
Re: CA-75 ... Revisited?
« on: October 20, 2016, 05:09:30 PM »
Greetings!
Well, ok , I guess - thanks for letting me know I got the colors wrong - :
but if the SS3900 has what you need, then why use this mic circuit design?
The mic is set for tonal response - not full broadcast quality.
If you take a moment and do a little figuring, the Mic Amp in this thread is very similar to those you'd already find in mini chassis like the TRC-453/465 as well as the PC-122 and Cobra 146 - there is only one thing missing in the design (two actually but their routing is what is affecting this response) and that is how the feedback portion of the circuit is constructed.
In the CA-75, the Feed back path uses the 18K and 103/223 Disc cap - on the EMITTER legs...DC to DC block - but audio imposed from output of one to the output of another - this is where that "Compression" effect comes into play on this design.
Compare this to a PC-122 or TRC-465 :
- the feedback path is from EMITTER of one - onto the BASE of another.
Note, in the link above, R143 is 100K and C107 is 100uF versus R 5 is 18K AND C5 (a 103/223 was used here in TWO different models one for "Noise Cancelling" while the other was a simple "Power Mike" so it's not mislabeled - just confusing due to the nature that it uses two different values in the same spot for different results) Disc cap has to be used IN SERIES and would be connected to the junction of the EMITTER leg of TR26 and R142. As a sidebar... please note that R139 is a 15K - refer back to the Mic amp in this thread - as used in this threads' design - as R 6.
Because of the type of configuration shown at the link above, the DC block on the bias for both transistors is not needed...compression is achieved thru the positive feedback path back to the base of the 1st stage amp - with DC bias tracking using the capacitor not just for audio but for DC bias value averaging.
So what I see in what you are trying to do, is a bit of overkill in regards to compression - at least that is how I'm reading this.
So as to not go beyond the scope of this thread - I'd like to lock this to keep the Mic Amp being discussed here - as per the mic element as a unit - from turning into something I'm not really sure if it can be applied to the method you suggest.
The SS3900 has it's share of faults in this - one being that the ICOM - (I believe) uses low-level modulated design and uses frequency mixing requirements to achieve that level of drive and performance - places the class of output - let alone the quality - more closer to Class A or Class AB in the TX chain than to use the SS3900's Class D AM regulator design or THEN uses class A to AB design for SSB mode of operation - which is a different beast altogether.
This may not be a fair comparison...I'm a little cautious to proceed - unless I know that this thread and the information within it - stays as a stand alone unit.
Which is unfortunate.
Because - the Mic Amp design in this thread is an Emitter to Emitter feedback design, not an Emitter to Base Feedback design which can give you far greater bandwidth than the Mic amp design this thread is about.
I see why and thanks for the verification that these AM regulators - once properly developed in both bias drive and parts value pairing together - can make a good wide-range audio driver for broadcast - but I'm worried that this mic amp will not suit your needs due to it's design.
Regards!
:+> Andy <+:
Re: CA-75 ... Revisited?
« on: October 20, 2016, 07:31:18 PM »
Greetings!
Ok, now I see what you really wanted to know - I know you asked this in the first post, but I was, and still am worried that this mic amp may not suit your needs.
C5 - 103 or 223 (0.01uF up to 0.022uF)
C6 - 150 pF
C10 - 102 (0.001uF)
C11 - 102 (0.001uF)
All are used around the transistors directly - and are not connected to ground or source - directly.
Frequency compensation and bypass - affecting gain - frequency dependent.
Since these values are quite small - they only affect the higher tonal range. To remove them (caps), which I don't recommend - gain would then rely solely on the resistor values.
All the parts listed you wanted to know about are designed to provide audio frequency feedback paths - so they are affecting overall gain of each stage - so, the resistor values alone would make the gain in each stage - too high - and although, bandwidth would be much wider, it comes at a price I learned about, in removing C6 - that "starvation" effect I mentioned earlier can quench or even kill audio signal and induce a squeal instead.
C6 - is the biggest culprit - and when you look back in my reference pages - I recommended (regrettably) this part could be removed and substituted with a 270K resistor - unfortunately I found out first hand that this level of resistance caused more squealing issues in radios with a separate ground return for audio - like the 148 or Grant XL - so in reworks I found that this value was better IN A LOWER RESISTANCE range - like 100K or even less - and kept the C6 in there to reduce that "starvation" effect because of the lack of/improperly set bias - handsets own bias and drive - it just tamped / damped down the gain, the benefit I got from it recovers a lot of bandwidth.
Additionally - we didn't cover the operators own changes internally to their own equipment added another variable that I had to work with in choosing these values.
One effort to clear this up was adjusting/trimming/changing the gain factor as well as the input level of the Mic Element - I discovered a lot of the distortion was caused by high-input level - it was finally reduced and proper capacitance was applied to drive the amp input correctly (matched) the C10/C11 and even C5 fell more into place and really could have been left alone - they didn't affect the overall performance in a normally driven (unmodified) CB because of the peak response curve the microphone would now have after it's modifications - more closely matched the frequency response the radio had built into it.
Without the CB Radios limiter, there would be another problem of too much high-tone roll off and perhaps better said "muddy mike sound" so to improve the bandwidth product and offer a greater range of control thru the mikes' own handset the additional 56K resistor added on the C6 foil side as well as keeping the 150pF cap - WAS NEEDED to remove a tonal artifact caused by the element within the handset. This factor was part of the gain, output adjustment range and overall speech performance improvement into radios that had "tightly filtered" or limited audio bandwidth.
Although not a true fault for the Microphone maker or the operators own installation, if the station or location wasn't properly shielded nor was it properly isolated from wind noises you can easily discover it's ability to pickup background level ( cardioid versus Omni-directional pickup ) noises and amplify it, together with the operators own voice, added problems with compression performance - too much of a good thing - usually is...
However, again lost in the posts from previous - the C10 and C11 could be removed AS LONG AS YOU DID NOT ADD ANY ELECTROLYTIC CAPACITANCE OR ANY OTHER LARGE VALUE CAPACITANCE across R4 or R11 - this kept DC bias and Amplifier stage gain for audio close to each other and didn't add any equalization or frequency dependent gain factors. In doing this, you then left the stage open for self oscillation in certain RF leakage prone areas that can generate an RF feedback loop that would kill the audio signal and left you with a squeal or no audio at all because of the bandwidth the circuit would have had without those components in there the transistors easily clipped/saturated from the stray pickup of everything - even RF - floating around the cab/shack can be thrown in there.
To keep the newer amplifier design stable for a larger range of installs as well as radios this mic could be used with I had to reduce or change a lot of values. Why? Because I never anticipated problems that would occur - I used it with my equipment and it worked - even ones that used a Quad op amp for mic/squelch/AGC and audio pre-amp - Midland 77-113/115 and 77-160 - this mike worked well with them. But I had encountered other operators having problems with these mikes so - what you're reading are the results and solutions in solving them.
Again; you are seeing a "boiled" down version - so when it comes to what you're looking for - I'm not sure you'll find it. I'm adding my notes I have on this circuit once I know what you are looking for in hopes that you can use it - I just don't want any further perdition done to this thread...this is for the benefit of the mic amp and those that would like to experiment and use it.
So you know, you ask questions that I feel need a base for a proper answer - so it requires me to post these "winded" messages so you would know the why and what I did - in making this thread.
Regards!
:+> Andy <+:
Re: CA-75 ... Revisited?
« on: October 21, 2016, 12:34:39 AM »
Greetings!
I need to add more ...
Came across some older Stereo Equipment manual with a schematic, I attached it (still a work in progress) to the previous post in an effort to demonstrate that the network used in this stereo - although an IC - used a Series / Parallel Resistor and Capacitor network and in the way they did it shows something I wanted to point out to you in the previous message about C6 (150pF) and my use of a 56K resistor paralleled across it but replace the 8.2K resistor R6 with a 15K resistor.
There is also the Inverted Input of the IC using a Series filter network to ground, on the input, that still prevents DC (comparator action) but provides an audio frequency response as part of the Equalization network to meet RIAA standards for a Turntable.
The previous post also shows the Original Discrete Values still in place - before modification so you can "revert" this device back to stock if needed to re-approach from a different selection of values.
Why is this important? Your needs to provide a good Emphasis for Bass and perhaps a little De-Emphasis for the Treble range.
This also demonstrates the dynamic effects of using active filter networks and how they interact with each other in both Equalization and Output level. Proper selection is important to achieve the Frequency Response as well as the expected output.
So it's attached to the previous post for your review - it's a work in progress - but I hope you can glean something out of all this to help you achieve what you want.
Regards!
:+> Andy <+:
CB Tricks - END
The above is a revised posting to help answer some issues with CA75 since the original link was established back in 2010 - now as of later revisions - much of the info above may have changed from their original - so be aware - but the link to the CA75 mic is still available - so use it as you wish.
