• You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.

Direct injection question.

Grogan

W9WDX Amateur Radio Club Member
Oct 1, 2011
1,133
343
93
Southern New Jersey
was wondering about direct injection or a radio that can be modded for 3K wide receive and transmit. What would be the best radio. Thank for your input.
 

3k isn't very wide. Are you wanting am or ssb? 4khz (8k wide AM) can sound very good if set up correctly. 5khz (10k wide AM) is kind of the point of diminishing returns. Not many people are hearing that wide and more doesn't help much anyway.

The 148 works pretty well but it will not transmit on ssb the way I've always done it. I believe we were pulling out c174 and injecting audio into the + hole....it's been a while. It will take a good amount of audio drive unless you run a low carrier. Anything over 2 watts is too much. The thing are only good for 8 to 10 watts pep before you start running out of headroom and sounding like poo.

I have had decent luck with the 3 diode ultramod circuit on the 148 and tr296. It was used on plate modulated rigs of the day to keep the negative peaks from hitting baseline and to protect the mod iron. You can't over use it. It's just a safety to keep the negative peaks in check. You'll still need a peak limiter in your audio chain, this is basically your modulation limiter.

For AM only thise asymod boards work very well. The board will run cooler with a cobra 25 or 29 with a bipolar final. SSB rigs and mosfets are biased and the old 25/29 with the bipolar is ran in class C and more efficient....no flames, this is not an amplifier, class C is kosher for high level modulation. The asymod a knock off of the motor mouth maul mauldulator. It will let you fine tune your modulation levels easier than the direct inject thing.

I would rather have an HF rig than a cb, especially if you want to use ssb. You connect your audio through an isolation transformer to the data port on the back. On modern rigs that port usually will give you enough bandwidth to sound really good but is still filtered. Direct injection or an asymod will pass 20k of audio or more. You have to roll off the high frequencies with your equalizer or build a filter so you don't splash on adjacent channels.

I spent a lot of time and money on this crap and now use an Apache Labs Anan. The software has more audio processing than you'll ever need and all it needs is a mic plugged into the front. They seem expensive but it would have been cheaper if I bought one to begin with. I don't regret buying the other stuff because I learned a lot.

I still use my outboard gear with the anan. After a very long time I finally got to where I was completely satisfied with the sound earlier this year. The 200 to 500hz range was not doing good things for my voice. Sorry for the long winded response. I need to get some radio time in this weekend. :LOL:
 
was wondering about direct injection or a radio that can be modded for 3K wide receive and transmit. What would be the best radio. Thank for your input.

You may know all this already - so I will warn the reader that this is posted for LEGACY reasons, not just specific...it's meant for others that follow this thread...I don't want to bore you if this doesn't suit you...

There are a lot of radios' you can try to do this to, but will they hold up to your scrutiny?

You ask two questions, and both deserve a lot to cover - but you don't have a lot of time, winters' are bad enough...you do have to shovel the driveway and go out to get mail and groceries every now and then....

One is - Direct Injection.

I'm presuming you're setting up a station with a better setup than a simple mic and cord hookup to a radio - you want a more dedicated system for it.

Well, are you using a computer to process the audio for transmission or are you making a system that is yours and not specifc to a software driven program?

I'm thinking direct - as in dedicated - directly into and specifically for a radio.

So to answer that, you need a little background of differences in the two most common types of radio modulation used...
  • One is Reactance modulation - that is simply a transformer that passes battery voltage thru a coil...
    (power supply) DC power thru it and after "about so many turns" into the coil winds of the transformer, they place a tap and add a capacitor to it - so they can "inject" amplified audio into the transformer. Then as the power from DC as well as Audio "melds" together thru these winds - a reactive transformation takes place where power is changed - some is converted to audio frequency via a change in phase relationship - yes, it kinda acts like AC - but uses audio frequencies. Eventually it's applied to the driver and final stages. To protect those stages a "spike diode" is used to keep the power DC and audio moving in a positive direction.
  • Another is AM Regulation...the audio is amplified thru a mic amp circuit as above, but instead of placing audio thru a transformer it is sent into a transistor amplifier used to generate this power using DC pre-set to a level of DC - I call BIAS - a level of power for the Final and Driver stages to produce a carrier - and audio is added in thru a transistor amplifier set up like a mixer - which uses, or looks at the pre-set level of BIAS as a zero -start-point. Meaning that the audio envelope and all volume starts from this DC level - volume can go up or down - and the amplifier attempts to keep DC BIAS stable (it uses a regulated voltage to set this reference point - AM PWR) but frequencies in the Audio signal are applied thru the mixer that induce a swing of power thru a range of frequencies - this is generally known as the "envelope" - it is a range of frequencies that the mixer ties in the BIAS power and sends it to the amplifier - arranged in a fashion called "Darlington" that takes battery or power supply DC voltage and converts this into that signal that powers the Driver and Final. In this type of circuit, Spike diodes are not normally needed for the DC BIAS present in the audio signal helps keep the Driver and Final forward biased - does not mean that it's fail proof, but it is not designed for a wide audio bandwidth nor a large swing in power at the DC BIAS level to affect power drive for the output stage - if too much is altered and no protection provided - well, it pretty much self-destructs.
If you're going direct injection - you can alter the mic amp to help work as a buffer stage to the audio amp - or you can bypass the mic amp and apply the audio directly.

Both have pluses and minuses - Mic amp can be tailored for you requirement of the 3kHz bandwidth by simple filter and bypass / feedback techniques, but it is not a cure-all for all. Some limitations are the power level as well as the distortion products and artifacts the amp can introduce to the signal because of it's own power and bandwidth limitations due to the drive level the amp will have to amplify and pass thru.

Direct coupling to the audio (or mic) amp is ok, but also remember there are Equalization limits as well as bandpass in, thru and onto the next stages of amplification.

Have you thought about Parametric Equalization? This method is a little more adjustable and allows you to tailor more for the needs of you to your radio. Not for the feint of heart though...

Graphic Equalization can work. But, you have "fixed" bandwidth and frequencies - and due to the nature of the filter beasts inside - they are not narrow enough and can expand the sound beyond the capabilities or your needs to make the system work like you want it too. They can be adjusted - but not able to trim off the outer edges sharply enough, which can again - introduce distortion and artifacts that color your audio and you may not like the results.

Parametric EQ systems run several filters in "bands" like Bass Treble and Midrange - they are the three basic building blocks for any type of EQ system. You can use one, but you limit yourself in several areas to tailor your effects as well as offset limitations to the radio or any number of radios you wish to use this on.

Not just 3 bands to play thru and amplify, but Parametrics' also have the ability to center in on a range of frequencies within that band to peak out and also adjust the "Q" or the sharpness of the filter as you peak in that range.

I went here,
https://buildyourownclone.com/products/paraeq

Wish I had this back in 2008 before the "Crash" when, I too, lost a lot of stuff I was just beginning to play with...

Secondly - receiver - if you use a CB radio - pretty much any CB (AM only are more preferred) would provide the bandwidth you seek, but quality varies - so purchase a good quality radio with several noise reducing features is better than a simple receiver - even though the simpler the receiver - the better that bandwidth is - but it comes at the price of higher noise levels too...

Now as far as radio - Receive? Pretty much AM-only types can fit this scenario well, some SSB types that have "conversion" in several stages before "deconversion" also can work. SSB - base radios have to compromise performance to bandwidth so they are pretty much ruled out for any true level of quality based upon the single or dual conversion principles. AM Regulation based injection provides the easier level playing field but Reactance modulation if performed correctly can inject a level of quality and punch in signals' drive that is hard to rival.

So what's your game?
 
The transformer based modulator in the 29 and the transistor modulator in the 148 have both been discussed. Both have the ability to sound excellent when properly modified but the transistor series pass modulator has more bandwidth capability than the transformer. The difference is small and mostly in the low end bass because the transformer will distort the lowest frequencies before the transistor.

Even the best modulation transformers struggle to get down to 20 cycles and many inexpensive ones start introducing distortion under 80 cycles. That's because the DC current feeding the final will saturate the core under the load of heavy low end bass. If we were to drive a series pass transistor modulator with a DC coupled AF amplifier, you could control the carrier level by the amount of DC injected in the mic line! Meaning the transistor modulator can be designed to have a bass response down to DC.

In fact, the series pass modulator used in the 148 does go down to DC and it does control the carrier level through that modulation transistor. The only difference here is that all of the audio stages in front of that Darlington transistor pair are AC coupled through a 1 uf electrolytic. Modifying the audio stages in front of that modulator and adding one band of passive, parametric equalization inside the radio (copy of old electric guitar tone control) to "fix" the 200 to 500 cycle range makes these rigs sound broadcast quality with just a good mic plugged in the front.
 
Last edited:
It took me a long time to figure out the 200 to 500hz range was not flattering. Eventhough everyone told me it sounded good but there was just something not quite right.

I have not attempted it with the cobra 29 but believe the low range distortion can be improved with a diode string between the mod transformer and the driver and final. That should lower the dc current through the xformer. Basically remove d8 and install more diodes in series like the old school swing mod. This will also lower the carrier level so you can only go so far.
 
Good info bump. I am trying to dig up information on modulation transformers for my Cobra 29LTD. I will start with the stock transformer to see if I can make it all work then would be interested in replacing it with a Jensen or something. Does anyone use an aftermarket transformer in their 29?
 
Myself? No.

But I have seen tried efforts...

Now they can work to a degree but to really think of a swap you'll need to know some things before you try this...

First...
Are you going to stick with the original Reactance Modulator or Make your own?

It is a simple series of winds, a tap, then more winds and another tap - and so on...ad nauseam...

You can arrange it any way you want, but follow schematic and take into consideration the board layout and your wire AWG size. You need to make it fit in the holes and slots provided...

Secondly...
Remember this is a coil, designed like a choke - for the efforts of pushing your voice thru, changing it into a miracle of swirling mayhem to arrive to a level of power and signal mashed into one single - er, mess on a wire...and others can then operate equipment similar to yours in decoding such information. By doing so, we all look forward to hearing from you as we gather around the warm glow of the S/RF meter lamp - waiting for the "weeds to wake up".

AMOnlyRadiosAudioTxfmr.jpg

It's not the best for wide banded broadcast, but in light of the narrower result - you do get to be "friends with benefits" and boost your audio by spiking voltages and phase changes re-directing voltage and current to make a simple hello into your microphone replete with frequencies scattering around the main center frequency and their voltage components adding a little extra power - as Fandom, er - Phantom.

So to compensate for that they re-equalized the Audio chip ; that being, more Trebel with less Bass - now they had to also remember the choke wants to remove the Trebel and convert it to Bass. So even the mid-range was going to shift lower too - so they had to place limits in order to pass anything thru the choke and not make it squeal.

Hence the 3-rd leg or that darn Non-inverting input bypass...
FeedbackOffsetOpAmp.png

I'll let you digest the above while I get the XYL happy again...
 
Ok, while I've got a minute or two (or hours) while the XYL naps...

Wanted to bring up the "3-rd leg" - the Other third of making Mic amps work like they do.

I guess I just also needed to put two sense, Cents - in for the purpose of making an effort to answer a question that comes up quite often.

Why can you just put the audio thru into the Mic wire and use it's own built in amp instead?

Well, you can, but you also have a problem with the Microphone Amp sections customized frequency response that favors the voice range at the expense of the bandwidth used for musical.

Believe me, if I could make this work I would, but it requires more of a bandwidth and stability fix that makes you use a Direct Inject method to obtain the desired frequency response and Equalization effects. The Direct to the Mic wire (Pin 2 Audio) at the Mic Jack can handel - er Handle - it...the Mic amp itself isn't designed for it.

I'll put this up and I'll address another problem dealing with AM-only types...a little later - because both small signal and large signal - we have to start somewhere - so small signal provides us some way into being introduced into larger scale signal processing...

InputOffset_4558Chip.jpg
The above is for your enlightenment - there are "boxed" and highlighted sections on each one of these Mic amp designs. The Box side, is the EMPHASIS circuit used on the differential amps other input to provide a frequency compensation curve that narrows and selects (filters) the tonal range most mic elements and voice coincide together on.

The Highlighted sections show the Mic Element has to pass into a circuit that BIASES the amp (Pin 3) within a given class of operation and with a given level in impedance for the signal to amplify. But only to provide enough impedance for signal level and stability to coincide with the LOUDNESS of the voice and limit Dynamic range by keeping the drive capacity low.

Coincide? Yes, for the same of definition. The Voice with all it's inflection, tonal disparity and resonance within itself - is unique to each individual but pretty much in within the same range of frequency bandwidth. So the Mic Element itself being only the medium to convert it to electrical signal. They chose elements that are economical and yet effective enough to provide vocal range but not much in fidelity. The filtering provides the mechanisms to limit drive power, but allow for some tonal range yet keep the audio frequency range narrow enough to accommodate the needs of the follower circuits and their input requirements.

They also chose the Bias level to keep the Amplifier linear but not make it unstable.

GalaxyUniden4558Chip.png

They also have to look at the GAIN factor of the amp to provide some linearity but also compression of the softer vocal passages, to within a ratio to the louder vocal sounds. You don't want too soft of audio for it can get lost in the required BIAS acceptance levels of drive the inputs of the following stages. Too loud, you get a lot of clipping distortion.

You also have the Bandpass limits. Now most of the Tonal range of voice we listen to is about 5kHz from about 80Hz onto as high as 5kHz - but that is when someone is talking to another within 3 feet.

You don't need all that range nor is it wise to produce that much signal bandwidth. Another factor is the filtering and processing that is required on the Receiver side, the listener - their radio can only do so much to provide the Fidelity.

That is where the two R/C components are used to provide COMPENSATION and levels of discernment for the listener. The other Resistor values are to provide the Amplifier input with some drive power to amplify the Mic Element Signal. Between the two - these values help provide the compression level and Dynamic range - too much dynamics will cause distortion - not enough compression, the softer passages get lost in Bias levels mixing in with and washing away the signal.

I had to show you the above, because they amplify small signal to branch into other circuits that - one, is frequency dependent not a lot of audio power is required - just compression, while another uses BIAS mixed in with the Audio to provide Envelope and Carrier power as well as Audio Drive.

WhatTheAmpMakes.jpg

With the understanding of the Above, remember that Galaxy, or any typical AM/FM/SSB radio (that uses AM Regulation) has to work within limits. So you can use Direct Injection on an AM signal and it will sound like a typical AM Radio station. FM does not use Direct Injection, but DEVIATION - too much audio Spectrum going in, it's going to cause pulsation and resonation (a type of tremolo effect) from intermixing and frequency shift of signal as a predominate spectral component can be generated in mixing products of Deviation. FM also has Dynamic range limitations - Volume is obtained thru carrier Power to deviation ratio changes, but the Audio tonal frequency information is in the deviation of the spectral range as a relationship to carrier frequency. SSB mode most of the spectrum fidelity is lost in the crystal filtering and IF mixing for Heterodyne up to your CB channel - and Frequency. SSB requires a limited drive level and too much dynamic range will cause saturation and other distortion effects.

Now, with the above in mind. We now have to look at taking this signal - whether we wide band or not - into stages that convert a high-current drive signal, with low-voltage - into a signal that now has a higher voltage range component and current is kept to DC for envelope purposes.


TDA2003AudioAmpSchematic.jpg


The above is a TDA2003 Audio Amp chip that replaces the 7222AP units. They are simpler but still offer the same basic designs the 7222AP used to amplify. So we have the basic building block to start with...

Cobra29IC4Gainloop.jpg

(Remember Q4 and R 7 referred above to the INTERNAL amp sub-sections)


Will get back to this as time permits...
 
Last edited:
Well, one of the many problems I've found in "Direct inject" is what the user really wants.

Some want broadcast quality audio.

Others simply want audio "punch" or dynamic range of signal.

So the above post is really just a simple primer for the next stage of questions the users should ask so I know where to go next.

Remember the issues around bandwidth - and the Vocal component you use in any typical radio.

IF you "broaden" the frequency spectrum, you lose the ability to provide the punch and dynamics of the vocals.

So, that also means - you can be a duck,. Or a mud - duck - either way, with the lesser power levels for fidelity versus intelligibility. You'll need power on the other end to help you get the words out...

In any sense, certain limitations have to be set - else the users going find themselves with a Worked All Hot Homeowners (WAHH) award...

To try some efforts, the Galaxy and any radio using the 4558 chip, just a few simple parts changed can broaden the audio wonderfully for AM and FM, but makes SSB sound weak and like CARPE-Diem-Do-Do.

So for proof, look for the R/C you have on that 3-rd leg of the 4558 chip. Change them to 1uF~2.2uF and 470~560 ohms and observe the results. SSB, it's a lost cause too much distortion is crammed into it. But in AM and FM, you're beginning to see results.

Get a Squeal, then back off that 270K feedback resistor - lower it's value down to 100K - this again is the intelligibility versus fidelity problem.

Not for the feint of heart...

So Direct Inject and skipping the above ENTIRETY of this post and going to the AM Regulator input cap (C299) is why most go this route - you lose FM - but you may need to re-think the injection site...at least as far as connection methods.

I'll get back to that soon enough.
 
Direct inject bump. Jus got my Madison back from Loose Cannon with direct inject. It's way different than going through the mic jack with no mods. I am going to get a bigger EQ so I can adjust the audio width a little better. I am also investigating the dj style cross overs to make an adjustable high cut and low cut. Figuring it all out is the fun part!
 
  • Like
Reactions: tecnicoloco

dxChat
Help Users
  • No one is chatting at the moment.
  • dxBot:
    Tucker442 has left the room.
  • @ BJ radionut:
    LIVE 10:00 AM EST :cool:
  • @ Charles Edwards:
    I'm looking for factory settings 1 through 59 for a AT 5555 n2 or AT500 M2 I only wrote down half the values feel like a idiot I need help will be appreciated