2SC2999 MOD ANALYSIS

[OldTech03]

I think in layman's terms what Andy is saying is that if a manufacturer obtains test results for a specific part in a circuit designed to generate the best results they can get and then you put that same part in another circuit that does not resemble the test circuit it is unlikely you will realize the results obtained in the test circuit. Wooosh that was not totally layman like.

 

[Handy Andy]

Correct.

But to define it; because along the way, the previous parts are never re-tested in the new test "jig" so no fair comparisons can be made and is why the Myth and Legend are now what we are dealing with - a Rumor Mill of sorts.

The original test circuit was replaced, but the parts of the original are not used in it for comparison.

So we are told to "upgrade" the RF amp with a new 2SC2999 (or thereabouts) because the customer wants it done. The radio we are to put this in, is not the same radio the part is in, we are doing this conversion to another platform.

If it doesn't work - the customer is not satisfied and we lose a sale.

What did we do wrong? We replaced the part - and in this radio - it didn't make any difference to the customers' point of view.

We are getting blamed for something externally that can't truly be applied to the current platform - we have to MODIFY that platform to accommodate the new part and it's support.

However, if we MODIFY - we no longer have a Brand X radio - we have a Hybridized radio using another circuit to obtain the new performance - if any.

This is why this is more subjective than applicable - nothing can be made definite because what was once considered a template has now changed and no longer a standardized platform for all the manufacturers' to follow.

There used to be SAE and Metric argument - now we have ISO standards, but it seems they are not applying it (ISO or any standard) equally amongst all parties involved in the manufacturing process.

So wild claims are made and I'm trying to demonstrate the use of one part using a set design yet it is installed in another platform idealized using a different set of performance parts based on another design that can be discerned as not one used in a normal parameter for the expected level of operation and performance.

...We put this mouse in this Hamster cage because the original squirrel retired...

Pretty much sums it up...

:+> Andy <+:

 

[Handy Andy]

There is something in this entire thread that stands out about myself in this ...

Original versus upgraded.

One side of this - people are demanding an upgrade because of the inherited traits the original design had, are no longer acceptable. Whether it's in power or modulation or receiver performance - many a customer asks us - would it help to keep the radio or should I get a new one?

Now to me, the original is what got them into radio.

If I brought out the 955 or a RCI - and then...

If I brought out a replacement CB just like what they had...

The customer has to choose - but asked me to help them make that choice.

Which one is fair - a direct replacement or an entirely new platform and layout with totally different technology.

I'm up against these choices every day...

I'd like to know what others think...

 

[unit_399]

Andy -
a new design based on today's technology would be welcome by a lot of us. Trouble is, the new radios are difficult to maintain and/or modify since none of them have a schematic available. Some of the rigs coming out of China are good performers, as is Uniden's 980ssb. But without a schematic it's practically impossible to evaluate the circuits and decide what changes to make to improve performance. Plus (without a schematic) repairing them when something goes wrong is like working in the dark. Not to mention parts availability.
So, I have to go with the older technology for now.

My hope is that when the sunspot cycle starts next year, interest will pick up enough to convince Uniden to bring out a couple of up-to-date exports (non type- accepted) complete with service manuals and schematics. Probably won't happen, but we can fantasize .

- 399

 

[unit_399]

I think in layman's terms what Andy is saying is that if a manufacturer obtains test results for a specific part in a circuit designed to generate the best results they can get and then you put that same part in another circuit that does not resemble the test circuit it is unlikely you will realize the results obtained in the test circuit. Wooosh that was not totally layman like.

Close enough.

- 399

 

[Martian]

The 980 schematic is out there.
I just don't want to see the guy accosted for it.

If by newer tech you mean SDR I doubt it happens.
Simply too expensive for a CB radio, which is a built to a price item.

I know of a group that is working on their own design. It's just not discussed openly.

 

[El Kabong]

I keep waiting for a Chinese company to come out with a cool, lowish priced general coverage HF radio the size of a CB that, with the cut of a diode, display's channels, scans, compressor and does 50 watts. If we can't get a quality CB the usual way maybe we can get one through the back door.

 

[Handy Andy]

To add this, because in a way - even the Test Bench stuff doesn't always show their cards either.

It deals with the power feeds into a test circuit...

Since the covers been blown on the fact that someone invented a part and die design that they feel is better than all the rest - but when placed in circuits prototyped with the various test jigs in mind - they fall flat. Because of direct trials in the real world they never did truly see forthcoming - we have empirical and physical as well as the FINANCIAL wasteland of evidence to prove they are not what they seem to be.

So dirty pool is played and suddenly eureka is shouted from the pulpits of a press meeting with some students and bystanders hired to say a few good things about this ...er - thing - and it works and works well (just don't ask them how - PLEASE!) Then they, in the Pubic Relations orifice - produce a PDF or two and offer free samples' of it - the test circuit, product support and ice cream to the first thousand vendors - whether they even sell transistors or not...nothing like free press...pressing on...

Not to bore you with repetitive diagrams, but this is something not too many people are really asking about...when it comes to the care and feeding of an Infant Transistor until they know how it will grow up and work out it's life as a functional block in society.

Power supply...

Yes, how do you feed a new part? Well, that requires a means to start out small and gain strength - usually with voltage and current steppers that allow the part to be examined and demonstrate it's stability in proper polarity of operation - including the die, wiring - stamping and epoxy along with the welds and leads being attached. Make sure it powers up and doesn't short right away...kinda' thing.

It may sound hilariously redundant but it is a quality of inception issue - you create the design and unless you see the physical model of it under test and passing all those trials. Variables exist, no matter what. Even the best model runs can't show every aspect - they just have to verify it won't blow up the table or create a quantum tunnel and suck all the air out of the room or introduce a gorilla of an unstable element from the place it contacted before we realized what it was doing and turned off the power to it.

Hi honey I'm home! What's for Dinner!?
GROWL! <SLASH>

"Next Week on Judge Judy...Paying the price for cheap groceries from a cheating cashier..."​

Did anyone ever think about how to keep it happy in an environment it's not?

When I see those jigs, I can't help but wonder how the power supply itself is designed to isolate itself from the circuit but yet act as the provider?

Since we're looking at variables - I can pretty much guess they use a variable power supply and we too, should design these concepts using such a means to isolate the part from potential problems in power feed paths. Referring back to those diagrams - usually they use a bypass cap - you see that in each and every one of those test circuits. A feed thru cap of some value is used to filter out the externals for the benefit of providing clean pure clear power for the device to show it's true nature.

But, did anyone else notice the base region? How it not only receives power, but the amount of power it receives - like an expected, inherited, impedance - like a reactive component high enough in value to take it out of the equation.

So I am bringing this up because many a shortcoming in todays radios can be attributed to poorly designed power feeds in amplifier stages that added more of a source of noise and unfettered feedback paths from the power supply rail or the adjacent circuits - and they had to scramble to "band-aid" the problem.

We can make the part operate linearly and use an R1 and R2 ladder divider circuit in some fashion to provide the power source but then are we isolating the part and the power source from each other?


Yes, look back at the test circuit used - and in which configuration was it used in.

So, if we have an idea of the required power in voltage and current in a DC realm, we need to understand that we also, are dealing with an RF spectrum issue and we don't want anything but the RF we want going thru it - to be amplified. The part can amplify RF - but which RF do you want it to amplify, in the spectrum - is the study question...

This is very important, for in the realm of zero drain (extremely high-impedance to power flow) inputs where they are/appearing as, highly capacitive - we need to be prudent in the design of a power supply feeding system that maintains voltage and current to allow a current to flow thru and or a power level window and provide the input for sourcing but then remove the power not needed thru return - we don't want the part taking in too much power it is not designed to work with.

A good example of this is the MOSFET - might as well accept the fact that we are going to be dealing with these in several ways in both power supply and power RF delivery.

So we see these RF transistors as having a very streamlined die design and small surface to apply a signal to the central region of the die - we also have to remember that the speed of light and it's relativistic realm in RF is a finite speed and we can introduce problems and unwanted artifacts if we do not isolate the power supply with a level of isolation to reduce the interaction of the feed into the part from the parts' needed interaction with the signal it's trying to get in there to amplify.

Anyone whom has access to a Galaxy radio or even a CB radio that is using IRF MOSFET's for the Final - can see resistor divider networks that isolate the Gate as a divider from turning on and staying on - to being able to amplify but little else from the SOURCE as well as Input - but reflecting back (SIC) we also inject a SECONDARY problem of a power flow that can hinder the level, quality and ability of the signal to arrive at the destination. They added in resistors into the bias feed a SERIES sourcing but the voltage divider feeds this voltage into a higher impedance device - so the device sees power at a given impedance thru it's Gate region and the Signal itself sees this inclusively as a high impedance admittance.

In a basic sense we need to review, reexamine, our designs and then properly isolate the feeds from each other to keep unwanted feedback paths from appearing. So it looks like we'll need to reevaluate the expected single point impedance and yet use a divider to allow the injected signal a path to exit the region as well.

Someone designed a transistor - thought it was the best thing since sliced bread - but, in the real world - it sucked. So they then designed a test jig to make that part shine and look like it was the RF worlds savior - because if you add in all those extra parts, it functions like never before experienced in real life.

Oh really? Said the Secretary - as she busied herself with typing up yesterdays cliff notes about the bathroom stall issue...

:+> Andy <+:

 

[Shadetree Mechanic]

Just found this again, bump.

 

[Robb]

Think that since there aren't enough techs to make proper receiver alignments, noobs got the idea that great receive is still accomplished with a magic bullet mod. Honestly? If you are running an AM only radio; it really won't make any real and noticeable difference. Sorry. Unless the radio's stock receive transistor hfe is under par or substandard, or the radio's receive alignment is out of whack. But that can also be true.
Either way, an alignment when installing this 'mod' still becomes necessary - IMO . . .

 

[Handy Andy]

Along with several physical changes to, additions thereof, the components and their location RF amp section...

I got flamed for "Reinventing the wheel" all because I stood up for Mike in his efforts to show Germanium diodes performance versus their newer Schottky replacements...people notice, so why not try to bump up the Receiver? Is anyone really going to care?

• I'd love to be the one that says yes this works.

But, that relies on the ears of the beholder. Their original parts are no longer made and the newer parts require support part changes to accommodate their frequency response performance and characteristics.
What one may hear is simply a higher frequency bandwidth the Schottky provides to the Audio Detector.
Is that noise a good thing? I don't know - Germanium offered a "latent" and slow response curve to frequency, as you increased the audio bandwidth the Germanium style Diodes simply rolled-off the high end tonal portion. Older folks notice the "Warm sound" versus the sharper - nearly clickety snapping sounds the audio and the noise it's embedded in, makes when it arrives to the listeners ear.

Others notice the AGC performance - pushes the noise floor down. Yes, it can generate that level or threshold power during detection because of the nature of the beast inside the receiver is trying to keep quiet, while the rest of the worlds own noises come roaring into the 1st Stage. So, yes you can call it a reception improvement.

The issue of this thread is about a Transistor - that can use these performance enhancements and it can and will do circles around most other radio equipment made in the day, for the modes you listen for and use still today. But that is not possible or needed for all radios or equipment - keeping the original design gave you predictable results as the engineers that built it, put in for those designs back in those days.

You may be asking for more trouble in trying to do the upgrade than to keep original design.​

No one, except the end user has to bear the burden of proof - after all that effort to Develop, Manufacture and Retail - no one bothers to ask the Customer "How does this sound?"

If you're the Older Guy, you'd say "It's a lot to listen to."

If you're the Younger Guy, you'd say "Give me more!"

I can see why Mike gets annoyed. He was trying to send up a flag.

Now, if you want that clickety sharp sounds of these newer radios - that's fine. But to attempt to retro-back this performance into the older radios requires more than the outcome may be worth. Sure you can fix the RF amp section and do up a card that is a drop-in for the original. Yes, you can make the AGC work better (relative term) - but why?

Only if the receiver needs help. Was it damaged? Then we go back to Robbs' question of the PIN diodes. They may be the biggest problem - because they suffer from the most abuse and damages caused by RF power spikes and noise pulse events that can serious erode the ability of the PIN Diode to stay sharp and work only when needed. They do degrade from impulse events and they tend to lessen the receivers ability to keep listening.

We need to remember that Schottky may cure a symptom, but the problem is still there. MC301's are hard to find.

They instead offer a new external speaker upgrade with tonal filtering to make it sound LIKE YOU ARE LISTENING TO THE ORIGINAL...

Sigh...

 

[Handy Andy]

Ok, I guess I need to just throw this out there...

And then we'd know...

If someone can make the 2SC2999 to 2SC1674 mod - on a chassis that uses the 2SC2999 transistor - then we may have something to see if such ideas actually work.

Meaning - find a chassis that uses 2SC2999 as the RF amp - then remove it and drop in a '1674 and then re-test.

Why "reverse" - to see if it's reciprocal. That's is, if the conditions of the test can be applied to the parts you wish to swap out - do they perform as well in a circuit design that supports the other part?

As far as "mods" for receiver performance...

A typical AM only radio - there is a lot to be said about it's own limitations in RX performance, receive sensitivity is one factor, but it's ability to DISCERN Signal from Noise - another, and between the two is the selectivity of signal.

Now some of you can ask, then why did you suggest a Ceramic swap out for a Xtal filter in the IF strip?

That is only if you need bandwidth for Oddballs - the above and below channels - the IF strip can tune the range but the Xtal was the limitation in the Receivers' performance.

Another factor in all of this is AGC performance...

Well, to be blunt - the AGC (in a typical AM-only CB) is really nothing more than a simple voltage divider circuit that is used to SUBTRACT or offset power potentials across the Emitters (or in some cases Collectors or the Bases) to affect the OUTPUT of the section. It's simply a Feedback loop - in principle of - power applied then reduced as signals become stronger in receivers' sections - so it prevents distortion and overload and allow Selectivity to overcome the Sensitivity the receiver has to obtain and DISCERN the signal from other factors in and around this frequency - being Noise and RF energy close to and around the frequency we're listening to.

• When you need sensitivity, the AGC line is powered thru 1/2 of this divider, it is from the POSITIVE side - it has battery voltage on the rail...so it powers all the sections needed to provide gain control in the strip to near full BIAS to obtain sensitivity - meaning it receives everything you're listening to on the frequency including the ambient NOISE present in and around the channel.
• When a signal or Noise is strong enough to be detected - it arrives at the Detector side of the radios IF output (Last stage) a diode used for the purpose of AGC starts to conduct - it's tapped on it's NEGATIVE (Cathode) end and develops a Negative Going voltage.
• This flows thru a resistor that is the other half (2 / 2 - or opposite side) of the AGC lines Voltage divider and it TAKES AWAY voltage from this rail that feeds power to this section of the strip.
• This Negative going power SUBTRACTS energy from the line, which reduces BIAS - and reduces the signal amplification ratios in the strip.
• Since it's inception is back at the detector - any result is dependent firstly upon the actual received signal(s) the RF amp gets - then is reapplied to reduce the gain of the stage and in some ways, provides a limitation as compression of dynamics the stage can provide. Again a factor of DISCERNMENT.

Also some radios apply AGC as an ability of a Amplifier to "sink" or remove power from the strip - so an OP-Amp is used as a means to reduce gain as a result of differential output applied to the AGC rail. You still have a voltage divider circuit - only now, you have an amplifier on one side used to remove (or apply) power to the rail to alter the gain of the strip.

With the above in mind, some limitations of the design, including the ability to react quickly to spikes and fast rise and fall signals - AGC can only do so much.

Some radiso only have "input limiter" as a pair of simple 1N4148 Small Signal Diodes to act like clipper and limtier functions - limits signal input strength and protects the RF amp, but you also get noisy results on strong signal inputs that force these diodes to rectify and conduct heavily.

So, you can use a PIN diode like a switch, or if Biased correctly - make it a variable attenuator that is dependent on both a Gain setting threshold, and the detected signal - into a BIAS that the Pin Diode can allow to flow across it's junction and either allow RF to pass, or remove RF as a given quantity to limit the power of the input before it goes into the RF amp.

So the PIN diode can provide a means to "regulate" gain by BIAS across itself can allow RF to flow into or out of a circuit (Pass or Shunt design) to further provide control on excessive noise and RF signals.

So with the above in mind.

AGC and PIN - controlling RF signals by strength - you can now see what I meant by focusing on the AGC detection and RF control - and keep the Schottky diodes OUT of the Audio Path - and detector that derives the audio. Keep the "recoloring" of audio to a minimum by maintaining original design parameters for the Audio path. You can allow the receiver to sound more like it's original design intended by keeping the original or as close to it as possible - in design and part support. You are just modifying the AGC and PIN sections to accommodate more demands placed on the receiver to DISCERN and SELECT the signals you want to hear.

When you change the RF Amp - you now are changing how the Receiver AMPLIFIES what is present Before DISCERNMENT and SELECTION takes place. So Noise and all other factors combined are now different so that puts you in a position of having to ALTER the performance of sections to obtain the best results in the Strip you've changed the performance of.

So if you're into redesign - here's the ball - run with it.

But if you're into preservation and only maintain for upgrade - then focus on the PIN Diode and Schottky FOR AGC and RF Gain (feedback principles and Amplifier gain limits) - keep these noisy parts from adding more noise and recolorizing your audio to the detriment of the receivers' own tonal quality and performance.

Ok, now with that being said, I've gotta get back to the life I know, and that means to take care of my XYL and to remind others to follow some SAGE advice...

Happy Wife = Happy Life
​

Regards!
:+> Andy <+:

 

[secret squirrel]

Bump, This thread answered a lot of question about the receiver mods.

 

[El Kabong]

After having Schottky diodes in my 2000 and seeing the higher noise floor for myself I declare the above correct. They're a rip.

 

[nomadradio]

Reminds me of a guy who worked at a CB shop in the back of a chrome shop in Winchester KY years ago.

His eyesight was too poor to do receiver alignment. But it was good enough to solder in the Workman preamp circuit board. That was his answer to receiver performance. Don't touch any adjustments, but install the preamp.

Had a sack of them we took out of dissatisfied customers' radios.

Shoulda written him a commission check for all the business we did as a result.

73