We seek to understand and document all radio transmissions, legal and otherwise, as part of the radio listening hobby. We do not encourage any radio operations contrary to regulations. Always consult with the appropriate authorities if you have questions concerning what is permissible in your locale.

Author Topic: Project 756 Pro  (Read 1412 times)

Offline Josh

  • DXing Phenomena
  • *******
  • Posts: 4322
    • View Profile
Project 756 Pro
« on: October 26, 2022, 2158 UTC »
Perusing QTH one day a few weeks ago, an advert touted a bespoke IC-756 PRO hf rig at a tantalising price. Sad to say it was spoken for, listed as sale pending, I pm'd the proprietor and asked to be put on "the list" if the deal fell through. Needless to say the deal fell through and a very clean almost new looking 756 Pro made its way to new ownership.
Total cost with shipping was $360 USD. Even if the tx section was gone forever the Pro series make for outstanding rx rigs, plus they have that neat fish finder! So $360 for a rig with a fish finder and the same dsp chip as the illustrious JRC NRD 545 was, to put it bluntly, a steal.

It seems this rig as delivered indeed had tx issues, and I immediately recalled similar issues with other Icom rigs from past experience, particularly the 746 series. Sad to say the same issue also affects the 756 Pro series. To wit, static or rf can impinge upon the rx grounding silicon innards (activated when in tx mode) and eat them. They're in line when the rig is off, and when an antenna is connected anything coming down that antenna can pop the doodads, due to the Icom's rx/tx switching scheme. After testing (apply rx max attenuation via the front panel button and if the tx is good with the rx attenuator enabled, well, you have the disease) this rig def had "the disease".

This is a link to a fix of the malady;
https://www.youtube.com/watch?v=EPnglEOLChE

So the rf section in question was investigated and bad diodes and a transistor were noted and replaced. In this condition the receiver section is active during tx and puts all sorts of garbage into the tx signal, it also attenuates the rx signal. This repair fixed the tx howling issue. Note the component placement and road map of the Pro differs slightly from the Pro 3 but you can figure it out as they do the same job. Enough about that, done and done, sorted.

But wait!

The original Pro flavour (subsequently remedied in the Pro2 and later rigs) was designed with another malady that often went unnoticed! Some early adopters of the OG (Original Gansta) Pro noted a weak audio trace of band activity present below the much stronger audio as delivered by the filter setting. In my case the rig would have several kc of the band tuned to present in the audio, this is best heard via headphone but it was present and hard to ignore once you knew what it was. There is a well founded fix for this;
"Correcting the IC756Pro Noise Blanker Drive-Signal Leakage Problem"
https://www.ab4oj.com/icom/756pro_nb/main.html
HA HA! Listen to the rumble! HA HA!

This is a simple fix consisting of a single lytic cap, and the dif is night and day, seriously. Once modded the rig has none, nada, zilch, zero, null, of the garbage present in the audio all the unmodded rigs have. Enough about that, done and done, sorted.

Then on to some mods to enhance longevity of Project 756 PRO.
The inputs of most modern hf rigs have some form of impulse suppression, a gas discharge thingy or two, most commonly the input to the rx section, and the PRO has such, but what about the dual UHF socket inputs? No protection offered to the tx section at those menacing sockets, so some gdts (gas discharge thingy) good for a bit over 100w were placed across the UHF sockets; center terminal to nearby ground.
https://www.littelfuse.com/products/gas-discharge-tubes/low-to-medium-surge-gdt.aspx
Enough about that, done and done, sorted.

Ok but what about the 12v dc input?
What about that?
What if the psu goes wild and delivers 20v or more?!?
Well, for that you employ a mov, a metal oxide varistor, a resistor that changes resistance when confronted by a voltage over a certain set value;
https://en.wikipedia.org/wiki/Varistor
I selected a 20v mov for that duty and now the PRO is about as well protected from external threats as can be cheaply afforded. Enough about that, done and done, sorted.


So in summary here we are with a running Icom IC-756 Pro on the cheap, and as you might have suspected, more mods are forthcoming!
We do not encourage any radio operations contrary to regulations.

Offline Larimar88

  • Newbie
  • *
  • Posts: 5
    • View Profile
    • Email
Re: Project 756 Pro
« Reply #1 on: October 28, 2022, 2113 UTC »
Excellent information. Thank you.

Offline Josh

  • DXing Phenomena
  • *******
  • Posts: 4322
    • View Profile
Re: Project 756 Pro
« Reply #2 on: November 01, 2022, 0154 UTC »
So glad everyone is enjoying the project!



As to further mods to bespoke 756 Pro, well, the pro series are already about as effective on hf as can be desired with the simple addition of that cap that eliminates the rf getting into the nb circuit. That said, there are some relatively easy and inexpensive mods one can perform that should give worthwhile results.

I considered editing the fw for the dsp, such as adding a synch detection mode akin to that of the JRC NRD 545 - as it uses the same dsp so it has been done, but that would entail a bit of watching bits blow by an i2c port (or whatever they use to get the rom bits to the dsp) in both rigs, and I don't have a 545 to watch bits with. Shame of it all.
Maybe someday!

Anyway, closer to reality are filter upgrades.

The rf section employs several 455kc Murata ceramic filters, not to obtain a nicely filtered output but to "roofing filter" the input to the adc, sans bespoke filters there may be 20kc (or more) of raw unadulterated hf bw impinging upon the poor adc. With the filters you get, say, 10kc or less, making the adc much happier. But that nice filtration also involves a digital world tradeoff as filters of the ceramic variety, nay, many filters used in radio circuits, introduce group delay, something digital circuits like adcs are better left without. In the analog world IF filters can have loads of group delay and a human listening to the output of those same filters will hardly notice it, hence their use. In the digital world, however, they are drek.
https://en.wikipedia.org/wiki/Mechanical_filter
https://en.wikipedia.org/wiki/Ceramic_resonator
https://circuitcellar.com/research-design-hub/group-delay-basics-more-filter-fun/
https://en.wikipedia.org/wiki/Gaussian_filter
https://en.wikipedia.org/wiki/Filter_(signal_processing)#Quartz_filters_and_piezoelectrics

At issue is an analog filter designed for best group delay will be pretty nasty from a traditional IF bw filtering standpoint; the skirts are going to be wide and get wider the further down one goes to the noise floor, and the amplitude response will also be all over the place going from one skirt of the filter to the other. A given filter designed for steep skirts will have highest gd at the edges, while a filter designed for best gd will have flattish gd but a big nose of amplitude right in the middle of the passband - also something an adc doesn't want to have to deal with.
Digital devices like adcs and dsps do not tolerate any disorderliness in their inputs, they demand flat amplitude and phase but analog filters can't do both at the same time.

Murata, conveniently enough, made some 455kc ceramic filters that have improved group delay as their selling point and they aren't very expensive either compared to their standard ceramics, so that's nice. Some of these reduced gd gems were even in the parts bin from a previous project, so why not use them? Not only are they better with regards to group delay they're also better in other regards, such as 6/60dB ratio (aka shape factor), ultimate stopband as well as leakage, they even come in a shiny metal shield can whereas the stockers are just resonating/radiating in the open in a black plastic box! So that is a mod that can be done with just a simple one for one replacement, however some circuitry mod such as short coax jumpers might be needed as they're not exactly pin for pin replacements.

Yet another mod, and this one is very easy as it is pin for pin replacement of standard parts, is a nb improvement.
Decades ago I had International Radio in Florida mod a Icom 751 to my specs, and it turned out swimmingly!
PIN diodes were used in the lower portion of the front end filters (reduces imd compared to PN diodes in the same spot) and Schottky diodes were used in the nb area. The nb circuit went from one where activating it with any but the weakest signals present would greatly distort the rx signal..... to one where during a local t storm you could turn the nb knob all the way up and literally make the lightning crashes disappear, and leave the underlying signals -loud or weak- undistorted.
It was an epiphany!
I'd love to have that nb in the 756 Pro, wich sadly is very lacking in the nb performance area, distorting most any signal when the nb is acting on a pulse.

To be honest I don't recall if Schottkys or PIN types where used in the nb H pad, but it was one or the other and I still have some of both on hand in the parts bin. What I figure is taking place when nb is activated and signals are being distorted, the sigs are being "semi rectified" by the pulse activated diodes, my guess is in the transition between fully on and fully off of the nb diodes during the nb pulse. Noise blanker diodes installed at the factory are in almost all cases simple PN types because they're cheap, not Schottkys or PINs wich are pricey. The good thing is most analog nb circuits are a simple H pad in any given hf rx; a few diodes activated by an pulse detector (usually a diode) and it looks like Icom copy/pasted the same nb design from the early days of Icom hf rigs to at least the Pro series.
https://en.wikipedia.org/wiki/Schottky_diode
https://en.wikipedia.org/wiki/PIN_diode
(fun fact; the common 1N4007 diode has a PIN-like structure)
Now for some testing!

So that's where Project 756 Pro stands today.
We do not encourage any radio operations contrary to regulations.

Offline Polar Bear

  • Jr. Member
  • **
  • Posts: 59
    • View Profile
Re: Project 756 Pro
« Reply #3 on: November 04, 2022, 2116 UTC »
If it is supported by the factory, a factory service center repair would be almost $500 for the one fix and as much as $200 for the other.
So by doing the repair by yourself, you saved almost $500
So $300 for a good donor radio - which you bought, was as much as what it was worth!
Kudo's for fixing it yourself.

ps. - ICOM builds junk, and instead of fixing their mistakes, being a man and saying I goofed and recalling a number of units, they do a really neat thing, they turn their back on their customers, change the model number , rectify the faulty part in the new model and keep selling it to anyone willing to buy it.

That is the only thing that holds me back from buying anything with the ICOM badge on it.

Offline Pigmeat

  • Marconi Class DXer
  • ********
  • Posts: 6684
    • View Profile
Re: Project 756 Pro
« Reply #4 on: November 13, 2022, 0616 UTC »
So, do you spot more walleye or smallmouth with it?

Offline Josh

  • DXing Phenomena
  • *******
  • Posts: 4322
    • View Profile
Re: Project 756 Pro
« Reply #5 on: November 13, 2022, 2242 UTC »
If it is supported by the factory, a factory service center repair would be almost $500 for the one fix and as much as $200 for the other.
So by doing the repair by yourself, you saved almost $500
So $300 for a good donor radio - which you bought, was as much as what it was worth!
Kudo's for fixing it yourself.

ps. - ICOM builds junk, and instead of fixing their mistakes, being a man and saying I goofed and recalling a number of units, they do a really neat thing, they turn their back on their customers, change the model number , rectify the faulty part in the new model and keep selling it to anyone willing to buy it.

That is the only thing that holds me back from buying anything with the ICOM badge on it.

I agree on the pertinent monetary facts as stated especially with regards to shipping to said repair facilities, yes a bundle was saved on a rig with the glorious fish finder.

So back in the day Icom decided to actually listen to a qst review, even replacing or modifying the rig qst tested and sent it back to qst, receiving a better review afterwards.
Then Icom produced the -A version of, you guessed it, the 751. And it was a fine rig.
Then they came out with the 761.
Then the 765 an upgraded 761 a year or so later.
Then the 746.
Then the 746 Pro.
Then came the 756, more or less a hybrid 751 with dsp and a fish finder.
Then the 756 Pro.
Then the 756 Pro 2 with audio fixes developed by HAMs for the Pro.
Then the 756 Pro 3, and that ended the line.... oh wait, ever looked at a 7600?
My guess is the 7600 is the latest version of the 756 Pro series.
Yazoo has followed Icom to a degree by making -A models of problematic rigs such as the 991 original flavor, now called the 991A.

On Project 756 Pro, looking into the appropriate diodes to replace the noise blanker pad, speccing parts is fun! Pretty sure the ones to start out with are gonna be schottkys as they are fast switching and low capacitance, but if they don't result in an improvement will go pin types. With before and after noise examples to follow.
We do not encourage any radio operations contrary to regulations.

Offline Josh

  • DXing Phenomena
  • *******
  • Posts: 4322
    • View Profile
Re: Project 756 Pro
« Reply #6 on: November 13, 2022, 2252 UTC »
So, do you spot more walleye or smallmouth with it?

They have to be bigger than the noise floor, but you knew that.
We do not encourage any radio operations contrary to regulations.