Topics

Software support Request for Full QSK BITX40

Gary O'Neil
 

Greetings Allard;
 
I've got my BITX40 modified to work Full QSK, but I've noticed your software doesn't support that mode. Any chance you can add this to your 1.27.x Code? I think your V2 code requires other hardware mods which I don't think are needed for the single band radios.The hardware switches in the sub or very low microsecond range, but its an unusable feature with your software due to the built in timer, pop mitigation delays, turning of CLK2 off and on, etc. Full QSK should be an easier mode to implement, since everything switches simultaneously (by necessity), and doesn't require sequencing. i.e. offset the display to read the transmit operating frequency, then simultaneously enable PTT and the CW Carrier. For full QSK, everything follows the keying. 

The fast switching enables it to work like a real SSB transceiver now also, but it is a bit problematic to use in Full QSK mode because it keeps the transmitter keyed between elements (Semi-QSK). The RF switch follows the keying during Full QSK so you can receive between elements; and since your software doesn't anticipate such instantaneous T to R transitions, the transmitter feeds back on itself for the duration of the hang time. while the key is up, creating a 5 to 10 W spur at around 8.5 MHz. LOL! My workaround for this was to use the Mic PTT line which doesn't introduce the delays. This requires that I force the carrier offset outside your code, and of course your code doesn't recognize this as CW, disables my ability to use the keyer features, interferes with using the rig on SSB, etc. No T/R or R/T transition pops are generated, so there's no need for generating delays to mitigate them. Of course the Hang-Time spur issue makes Semi-QSK unusable as implemented. 

If you don't feel quite up to adding another feature to the V1 code, can you recommend an earlier version I could migrate back to that preceded the CW upgrades?

72

Gary, N3GO


Allard PE1NWL
 

On Fri, Jan 19, 2018 at 09:34 pm, Gary O'Neil wrote:
I've got my BITX40 modified to work Full QSK, but I've noticed your software doesn't support that mode. Any chance you can add this to your 1.27.x Code?
Hi Gary,
In order to better understand your request, can you explain what modifications you did to your BitX40 so that it can do full QSK?
73 Allard PE1NWL

Gary O'Neil
 

Hi Allard;

I replaced the relays to switch and purge the TX/RX supply lines, and a SPDT RF switch to isolate the receiver during Transmit. It all switches in less than a microsecond, but with all the capacitive loading it has to dump, it's probably moving up to somthing close to a millisecond. QSK speeds are not an issue. The transmitter transmits instantly, and the receiver recovers to full sensitivity between dits at between 40 and 50 WPM (measured it at 48 WPM with my keyer at max speed), but I rarely operate at 30, and 35 is approaching my maximum copying speed.

72

Gary, N3GO

K9HZ <bill@...>
 

Wow Gary you are gonna have to share that so i don't have to reinvent the wheel.  No interest in QSK but i am interested in getting rid of the clunky relays.  The front end/PA isolation can be done with a regular set of diodes i think because its low power. 


Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ

 

Owner - Operator

Big Signal Ranch – K9ZC

Staunton, Illinois

 

Owner – Operator

Villa Grand Piton - J68HZ

Soufriere, St. Lucia W.I.

Rent it: www.VillaGrandPiton.com


email:  bill@...

 


On Jan 20, 2018, at 8:42 AM, Gary O'Neil <n3go@...> wrote:

Hi Allard;

I replaced the relays to switch and purge the TX/RX supply lines, and a SPDT RF switch to isolate the receiver during Transmit. It all switches in less than a microsecond, but with all the capacitive loading it has to dump, it's probably moving up to somthing close to a millisecond. QSK speeds are not an issue. The transmitter transmits instantly, and the receiver recovers to full sensitivity between dits at between 40 and 50 WPM (measured it at 48 WPM with my keyer at max speed), but I rarely operate at 30, and 35 is approaching my maximum copying speed.

72

Gary, N3GO

Allard PE1NWL
 

Thanks Gary,
interesting!

My most important development criteria is that the software shall always
work, also on a stock (non-modified) BitX40. Any hardware modifications
supported by the software shall be minor, simple, easily reproducable and,
most importantly, optional.

The modifications you did are pretty major, and I'm afraid it would be
quite a challenge to revise the code so that it supports both modified as
well as non-modified boards.

73 Allard PE1NWL

On Sat, January 20, 2018 14:42, Gary O'Neil wrote:
Hi Allard;

I replaced the relays to switch and purge the TX/RX supply lines, and a
SPDT RF switch to isolate the receiver during Transmit. It all switches in
less than a microsecond, but with all the capacitive loading it has to
dump, it's probably moving up to somthing close to a millisecond. QSK
speeds are not an issue. The transmitter transmits instantly, and the
receiver recovers to full sensitivity between dits at between 40 and 50
WPM (measured it at 48 WPM with my keyer at max speed), but I rarely
operate at 30, and 35 is approaching my maximum copying speed.

72

Gary, N3GO

MVS Sarma
 

perhaps next only to0 replicating human barain, in liter vein, Lol

Regards
MVS Sarma
 

On Sat, Jan 20, 2018 at 10:04 PM, Allard PE1NWL <pe1nwl@...> wrote:
Thanks Gary,
interesting!

My most important development criteria is that the software shall always
work, also on a stock (non-modified) BitX40. Any hardware modifications
supported by the software shall be minor, simple, easily reproducable and,
most importantly, optional.

The modifications you did are pretty major, and I'm afraid it would be
quite a challenge to revise the code so that it supports both modified as
well as non-modified boards.

73 Allard PE1NWL



On Sat, January 20, 2018 14:42, Gary O'Neil wrote:
> Hi Allard;
>
> I replaced the relays to switch and purge the TX/RX supply lines, and a
> SPDT RF switch to isolate the receiver during Transmit. It all switches in
> less than a microsecond, but with all the capacitive loading it has to
> dump, it's probably moving up to somthing close to a millisecond. QSK
> speeds are not an issue. The transmitter transmits instantly, and the
> receiver recovers to full sensitivity between dits at between 40 and 50
> WPM (measured it at 48 WPM with my keyer at max speed), but I rarely
> operate at 30, and 35 is approaching my maximum copying speed.
>
> 72
>
> Gary, N3GO
>





Gordon Gibby <ggibby@...>
 

Gary, if you would produce some sort of schematics and we can tell what’s going on, I am certain that some folks here can modify code to fit with that. Allard probably can’t release it into the wild, but we could get a version going that would work specifically for such a rig.

Sent from my iPhone

On Jan 20, 2018, at 11:42, Mvs Sarma <mvssarma@...> wrote:

perhaps next only to0 replicating human barain, in liter vein, Lol

Regards
MVS Sarma
 

On Sat, Jan 20, 2018 at 10:04 PM, Allard PE1NWL <pe1nwl@...> wrote:
Thanks Gary,
interesting!

My most important development criteria is that the software shall always
work, also on a stock (non-modified) BitX40. Any hardware modifications
supported by the software shall be minor, simple, easily reproducable and,
most importantly, optional.

The modifications you did are pretty major, and I'm afraid it would be
quite a challenge to revise the code so that it supports both modified as
well as non-modified boards.

73 Allard PE1NWL



On Sat, January 20, 2018 14:42, Gary O'Neil wrote:
> Hi Allard;
>
> I replaced the relays to switch and purge the TX/RX supply lines, and a
> SPDT RF switch to isolate the receiver during Transmit. It all switches in
> less than a microsecond, but with all the capacitive loading it has to
> dump, it's probably moving up to somthing close to a millisecond. QSK
> speeds are not an issue. The transmitter transmits instantly, and the
> receiver recovers to full sensitivity between dits at between 40 and 50
> WPM (measured it at 48 WPM with my keyer at max speed), but I rarely
> operate at 30, and 35 is approaching my maximum copying speed.
>
> 72
>
> Gary, N3GO
>





Gordon Gibby <ggibby@...>
 

My first ham radio transceiver was a full break in HW 16 from Heathkit.

Sent from my iPhone

On Jan 20, 2018, at 13:00, Gordon Gibby <ggibby@...> wrote:

Gary, if you would produce some sort of schematics and we can tell what’s going on, I am certain that some folks here can modify code to fit with that. Allard probably can’t release it into the wild, but we could get a version going that would work specifically for such a rig.

Sent from my iPhone

On Jan 20, 2018, at 11:42, Mvs Sarma <mvssarma@...> wrote:

perhaps next only to0 replicating human barain, in liter vein, Lol

Regards
MVS Sarma
 

On Sat, Jan 20, 2018 at 10:04 PM, Allard PE1NWL <pe1nwl@...> wrote:
Thanks Gary,
interesting!

My most important development criteria is that the software shall always
work, also on a stock (non-modified) BitX40. Any hardware modifications
supported by the software shall be minor, simple, easily reproducable and,
most importantly, optional.

The modifications you did are pretty major, and I'm afraid it would be
quite a challenge to revise the code so that it supports both modified as
well as non-modified boards.

73 Allard PE1NWL



On Sat, January 20, 2018 14:42, Gary O'Neil wrote:
> Hi Allard;
>
> I replaced the relays to switch and purge the TX/RX supply lines, and a
> SPDT RF switch to isolate the receiver during Transmit. It all switches in
> less than a microsecond, but with all the capacitive loading it has to
> dump, it's probably moving up to somthing close to a millisecond. QSK
> speeds are not an issue. The transmitter transmits instantly, and the
> receiver recovers to full sensitivity between dits at between 40 and 50
> WPM (measured it at 48 WPM with my keyer at max speed), but I rarely
> operate at 30, and 35 is approaching my maximum copying speed.
>
> 72
>
> Gary, N3GO
>





Gary O'Neil
 

Hi again Allard;

 

I don’t think the revisions I’ve made would actually be considered major. Aside from it switching too fast to be compatible with the manner in which semi QSK is rendered, it remains completely compatible with your code. I’m not suggesting that there’s a need to alter what you have however. I simply made the observation that Full QSK is not supported, and it should be a relatively simple and benign feature to implement... at least for use with external keyers. No hardware changes are required to remain 100% compatible with legacy implementations. 

 

This does expose a bug in your Semi QSK implementation however, and correcting that would make it compatible with Full QSK enabled radios. This is a paradox though, since the root cause of the bug is in the hardware, and your software cleverly masks it. We are thus left with the situation where two wrongs made it right. :-) Fixing it is not an option of course as it would disable CW in the legacy radios.

 

I also perceive a Full QSK feature with an optional hang timer set to zero (disabled) as easy to implement, making both modes (Full and Semi) available to those who enable their radios for full QSK.

 

Given the caveat that a Full QSK feature (and possibly the keyer) in your software is rendered useless without “additional” modification, everything as coded and wired today is unaltered. To wit... 

 

A1 is the assigned straight key input, and this is what Bugs and external keyers currently employ. No change here. You have already worked out the display correction for CW mode. Again, no change. All that remains is making D6 (CW Carrier) and D7 (TxRx) simultaneously track the state of A1 immediately without delay. Viola! Add it as a CW setup menu item and you’re done. Adding a delay variable as a setup option (default = 0) would be icing on the cake. 

 

You’re current code must remain intact to minimize the relay toggling, and I support your need and desire on the points you made on this, but I think this change might encourage others to innovate. What started out as a sorting exercise for patches to unresolved issues; led me to dig a bit deeper than I intended. When I discovered the issues remain after 10 years of patching, I felt it might be easier and less costly to mitigate the problems at the source. 

 

72

 

Gary, N3GO

Gary O'Neil
 

Hi Bill;

There’s limited interest in CW these days much less QSK. I’m off on a winter hiatus for a couple weeks or so, and was hoping to inspire Allard to hammer out a new Full QSK feature and have it awaiting my return. :-) It looks like I may need to get a bit more creative and wean the design off some of its software dependencies. 

I became a Navy radioman out of high school, and still pine for the return of those days when working Full QSK was standard operating procedure. I’ve not met any hams yet that actually use it though... not even ex radiomen (including myself). LOL! Many never actually learned how, and many hated having to even use CW if you can imagine that. I spent my entire time at Sea doing nothing but pounding brass while hurling into a bucket. Bittersweet times you might say. 

It sounds like you might be chasing the “Pop Pop Thump Thump” problem with its myriad of patches that almost work.  That’s what I actually set out to solve, but I really wanted Full QSK, and once I convinced myself that fixing those issues wasn’t going to lead me where I wanted to be, I decided to do the whole enchilada and make the problems go away on their own. 

I don’t want to publish anything that I have to defend because it breaks what folks have already done. I’m not compelled to purchase another BITX to verify what might work for you or others, but I can’t come up with a reason (other than the horrible switching speed) why it wouldn’t work with relays as well, and the speed isn’t an issue on SSB. 

You’re welcome to give it a try though, and I’m pretty sure I floated the idea shortly after I got mine running. I never heard whether or not it was ever tried and confirmed, but I’ve discovered that finding stuff in the threads of this group is daunting. 

Simply put, the root cause of those noises is stored supply energy, and they are easily mitigated by simply purging the supplies when they are switched off. It takes just 2 NPN transistors (2N2222/2N3904 or equivalent), and a pair of 2.2k resistors. Emitters of both go to ground, and the collectors go to each of the switched DC rails. The resistors are then connected between the base of each of the transistors, and the collector of the other. 

The the way it works is straight forward. When the transmitter goes active, it bias the transistor across the RX supply which dumps any residual RX supply voltage to ground. The opposite occurs when switching from transmit to receive, and the transmitter power rail is shorted out in the same manner. The 2.2k resistors limit the base current to around 6 ma to protect it from overheating the base-emitter junctions, while turning the transistors on hard enough to create a healthy short circuit across each of the supply rails, but only when they’re no longer being used. 

If no supply voltage. Exists (0 volts) on the transmitter, it can’t interfere with the receiver and vice versa. This works for very fast QSK, and since the relays take 7 ms each to switch, (around 15 ms total), the relay clicks will be the only switching artifact you will hear. 

No guarantees though. I’m not certain how close your radio resembles mine, and other issues may arise out of coexisting patch fixes.

BTW... Diode switches are not QRP friendly. Think 60ma or more. The switch I use consumes less than 100uA. Relays aren’t too friendly either, but they only consume power when transmitting. 

72

Gary, N3GO
 

Gary O'Neil
 

Hi Gordon;

I’m away from the bench for a couple of weeks or so, and I have to sort out some refinements, like maybe a work around for the current software. My hope was that Allard might save me that task, and I was planning to spend my time learning Kikad so I can do a board layout to make it aesthetically appealing. Most if not all of the interface connections can be made via the K1 relay footprint which will accommodate a standard 16 pin DIP socket. That would make it easy to replace the relays with sockets, and simply plug in a daughter board. The Raduino hacks are a rats nest of wires, and I’d like to clean them up to make the radio more reliable and field worthy.

Its not ready for prime time on its own, and I don’t want to spend more time fielding questions I’m not yet ready to address. It needs to work seamlessly with Allard’s software or it will be a headache dealing with folks that don’t quite understand how it has to be finessed in order to work. I can probably work through Allard’s code and get it working well enough to get the data I need on performance to test and refine it appropriately. I think all that’s required is figuring out how to disable all of the delays without causing it to branch off somewhere incorrectly. 

I think to make it work with Allard’s code as is will make it too complex of a mod to be of any interest. Its nothing more than a relay replacement, that doesn’t like the transmitter transmitting while it should be receiving. I could modify it to operate Semi QSK to make it compatible with the software, but what’s the point?

72

Gary, N3GO

 

Gary,

What did you do about the TX final filter relays? they click every time the PTT is pressed.

I have disconnected the line TX 12V line from the three filter relays next to K3 and directly wired 12V from C76 at the audio amp.
Now the relays stay on BUT the RX power draw has gone up.

Do share your mods.. I am interested in QSK!

Raj, vu2zap

At 20-01-2018, you wrote:
Hi Allard;

I replaced the relays to switch and purge the TX/RX supply lines, and a SPDT RF switch to isolate the receiver during Transmit. It all switches in less than a microsecond, but with all the capacitive loading it has to dump, it's probably moving up to somthing close to a millisecond. QSK speeds are not an issue. The transmitter transmits instantly, and the receiver recovers to full sensitivity between dits at between 40 and 50 WPM (measured it at 48 WPM with my keyer at max speed), but I rarely operate at 30, and 35 is approaching my maximum copying speed.

72

Gary, N3GO

 

Sorry, what I am referring to is the uBitx!


At 21-01-2018, you wrote:
Gary,

What did you do about the TX final filter relays? they click every time the PTT is pressed.

I have disconnected the line TX 12V line from the three filter relays next to K3 and directly wired 12V from C76 at the audio amp.
Now the relays stay on BUT the RX power draw has gone up.

Do share your mods.. I am interested in QSK!

Raj, vu2zap

At 20-01-2018, you wrote:
>Hi Allard;
>
>I replaced the relays to switch and purge the TX/RX supply lines, and a SPDT RF switch to isolate the receiver during Transmit. It all switches in less than a microsecond, but with all the capacitive loading it has to dump, it's probably moving up to somthing close to a millisecond. QSK speeds are not an issue. The transmitter transmits instantly, and the receiver recovers to full sensitivity between dits at between 40 and 50 WPM (measured it at 48 WPM with my keyer at max speed), but I rarely operate at 30, and 35 is approaching my maximum copying speed.
>
>72
>
>Gary, N3GO

Allard PE1NWL
 

On Sat, Jan 20, 2018 at 07:48 pm, Gary O'Neil wrote:
A1 is the assigned straight key input, and this is what Bugs and external keyers currently employ. No change here. You have already worked out the display correction for CW mode. Again, no change. All that remains is making D6 (CW Carrier) and D7 (TxRx) simultaneously track the state of A1 immediately without delay. Viola! Add it as a CW setup menu item and you’re done. Adding a delay variable as a setup option (default = 0) would be icing on the cake. 
OK, I get your point now. Perhaps it is doable. I can try to fix this, but there is always a risk that it interferes with other functions (because of possible timing/delay constraints), however I'm unable to test this on my own radio.
So I could try to do something and send it to you off group so that you can test it on your radio.

73 Allard PE1NWL

Gary O'Neil
 

Sounds fair Allard;

im off bench until Mid Feb, so no rush.

72

Gary, N3GO

K9HZ <bill@...>
 

Now thats a great place for a latching relay if you must use a relay.  


Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ

 

Owner - Operator

Big Signal Ranch – K9ZC

Staunton, Illinois

 

Owner – Operator

Villa Grand Piton - J68HZ

Soufriere, St. Lucia W.I.

Rent it: www.VillaGrandPiton.com


email:  bill@...

 


On Jan 21, 2018, at 4:34 AM, Raj vu2zap <rajendrakumargg@...> wrote:

Gary,

What did you do about the TX final filter relays? they click every time the PTT is pressed.

I have disconnected the line TX 12V line from the three filter relays next to K3 and directly wired 12V from C76 at the audio amp.
Now the relays stay on BUT the RX power draw has gone up.

Do share your mods.. I am interested in QSK!

Raj, vu2zap

At 20-01-2018, you wrote:
Hi Allard;

I replaced the relays to switch and purge the TX/RX supply lines, and a SPDT RF switch to isolate the receiver during Transmit. It all switches in less than a microsecond, but with all the capacitive loading it has to dump, it's probably moving up to somthing close to a millisecond. QSK speeds are not an issue. The transmitter transmits instantly, and the receiver recovers to full sensitivity between dits at between 40 and 50 WPM (measured it at 48 WPM with my keyer at max speed), but I rarely operate at 30, and 35 is approaching my maximum copying speed.

72

Gary, N3GO




K9HZ <bill@...>
 

We were thinking the same thing... shorting the opposite rail to ground of the mode you are in.  My circuit was a bit different using a modified schmitt to the base of the transistors but it provides the same result (that way one is not on while one is off at any time).   And i know i can design a diode switch that takes less current than the relay coil so good there.  

Im a CW op too for close to 50 years ... in major contests too but I never got the QSK thing. Seemed like a telephone call in the middle of another call to me but thats just me.  It would probably work for my wife who thinks much differently than me. 


Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ PJ2/K9HZ

 

Owner - Operator

Big Signal Ranch – K9ZC

Staunton, Illinois

 

Owner – Operator

Villa Grand Piton - J68HZ

Soufriere, St. Lucia W.I.

Rent it: www.VillaGrandPiton.com


email:  bill@...

 


On Jan 21, 2018, at 1:40 AM, Gary O'Neil <n3go@...> wrote:

Hi Bill;

There’s limited interest in CW these days much less QSK. I’m off on a winter hiatus for a couple weeks or so, and was hoping to inspire Allard to hammer out a new Full QSK feature and have it awaiting my return. :-) It looks like I may need to get a bit more creative and wean the design off some of its software dependencies. 

I became a Navy radioman out of high school, and still pine for the return of those days when working Full QSK was standard operating procedure. I’ve not met any hams yet that actually use it though... not even ex radiomen (including myself). LOL! Many never actually learned how, and many hated having to even use CW if you can imagine that. I spent my entire time at Sea doing nothing but pounding brass while hurling into a bucket. Bittersweet times you might say. 

It sounds like you might be chasing the “Pop Pop Thump Thump” problem with its myriad of patches that almost work.  That’s what I actually set out to solve, but I really wanted Full QSK, and once I convinced myself that fixing those issues wasn’t going to lead me where I wanted to be, I decided to do the whole enchilada and make the problems go away on their own. 

I don’t want to publish anything that I have to defend because it breaks what folks have already done. I’m not compelled to purchase another BITX to verify what might work for you or others, but I can’t come up with a reason (other than the horrible switching speed) why it wouldn’t work with relays as well, and the speed isn’t an issue on SSB. 

You’re welcome to give it a try though, and I’m pretty sure I floated the idea shortly after I got mine running. I never heard whether or not it was ever tried and confirmed, but I’ve discovered that finding stuff in the threads of this group is daunting. 

Simply put, the root cause of those noises is stored supply energy, and they are easily mitigated by simply purging the supplies when they are switched off. It takes just 2 NPN transistors (2N2222/2N3904 or equivalent), and a pair of 2.2k resistors. Emitters of both go to ground, and the collectors go to each of the switched DC rails. The resistors are then connected between the base of each of the transistors, and the collector of the other. 

The the way it works is straight forward. When the transmitter goes active, it bias the transistor across the RX supply which dumps any residual RX supply voltage to ground. The opposite occurs when switching from transmit to receive, and the transmitter power rail is shorted out in the same manner. The 2.2k resistors limit the base current to around 6 ma to protect it from overheating the base-emitter junctions, while turning the transistors on hard enough to create a healthy short circuit across each of the supply rails, but only when they’re no longer being used. 

If no supply voltage. Exists (0 volts) on the transmitter, it can’t interfere with the receiver and vice versa. This works for very fast QSK, and since the relays take 7 ms each to switch, (around 15 ms total), the relay clicks will be the only switching artifact you will hear. 

No guarantees though. I’m not certain how close your radio resembles mine, and other issues may arise out of coexisting patch fixes.

BTW... Diode switches are not QRP friendly. Think 60ma or more. The switch I use consumes less than 100uA. Relays aren’t too friendly either, but they only consume power when transmitting. 

72

Gary, N3GO
 

Tim Gorman
 

Bill,

QSK while ragchewing or contesting is pretty much of a waste. Semi-qsk
is fine for those.

But I used to use QSK all the time when I was traffic handling on cw.
You didn't need to send a whole string of characters for a fill when
the other end didn't copy. The other end could just break in
immediately after the missed character(s).

Been a ham for 54 years and did a lot of cw traffic back in the day. It
kind of died off about 20 years ago and I don't need it any more.
Semi-qsk with the modern radios seems to work fine for me.

tim ab0wr

On Sun, 21 Jan 2018 10:47:02 -0400
"K9HZ" <@Doc_Bill> wrote:

We were thinking the same thing... shorting the opposite rail to
ground of the mode you are in. My circuit was a bit different using
a modified schmitt to the base of the transistors but it provides the
same result (that way one is not on while one is off at any time).
And i know i can design a diode switch that takes less current than
the relay coil so good there.

Im a CW op too for close to 50 years ... in major contests too but I
never got the QSK thing. Seemed like a telephone call in the middle
of another call to me but thats just me. It would probably work for
my wife who thinks much differently than me.


Dr. William J. Schmidt - K9HZ J68HZ 8P6HK ZF2HZ PJ4/K9HZ VP5/K9HZ
PJ2/K9HZ
Owner - Operator
Big Signal Ranch – K9ZC
Staunton, Illinois

Owner – Operator
Villa Grand Piton - J68HZ
Soufriere, St. Lucia W.I.
Rent it: www.VillaGrandPiton.com

email: @Doc_Bill

Gary O'Neil
 

Here  ya go as best I can do on the road. I hope anybody that does this with the relays in their radio posts their findings. I would like to bless it with my  approval, but I won’t do this without having tested it for hidden demons that might arise in a contact bouncing environment. Resistor values are not critical and only serve as base current limiters. This should solve the self interference problem in all single channel systems of similar design, and is an absolute requirement to achieve Full QSK performance. If better understanding is needed, or important enough to optimize; you will need to 1) wait for me to test, document, and publish it, or 2) work through it on your own... a faster and fairly simple and basic task.

72

Gary, N3GO



7

Gary O'Neil
 

It appears my dwg failed to post. Will attempt ago when I settle in at campground.

Gary, N3GO