Jack Purdum

Arv:

All tough questions, but worth answering. Think of the possibilities. Personally, I think this should be of the free-to-use license, mainly because I don't want to see extensions to the bus that aren't under someone's control. Otherwise, the standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I had on my desk when I was the department chairman:

??? For God so loved the world, he didn't send a committee.

Still, there has to be a small knot of knowledgeable people who know the EE and software side of this. It needs to be small because it needs to be nimble, yet with enough technical depth to make things work. I've seen the agony of trying to define a standard (i.e., the X3J11 committee to write the first standard for the C language)...it is a formidable task, and the difficult increases with the size of the committee. There are all kinds of technical details to think about and the Atlas bus would at least be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE

On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <arvid.evans@...> wrote:

If hams were to design and standardize a ham-bus-system approach to modular equipment,

should the design be placed in public domain, or under one of the free-to-use licenses???

Should a group be formed specifically for the purpose of bus design, documentation, and
publication?

How would upgrades, modifications, and alternative bus designs be handled?

Seems there are lots of questions, lots of possible opinions, and lots of work to do.

Arv

Steve Black

#52616

开云体育

A camel is a horse designed by committee. Steve kb1chu

On 06/20/2018 10:50 PM, Jack Purdum via Groups.Io wrote:

Show quoted text

Arv:

All tough questions, but worth answering. Think of the possibilities. Personally, I think this should be of the free-to-use license, mainly because I don't want to see extensions to the bus that aren't under someone's control. Otherwise, the standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I had on my desk when I was the department chairman:

??? For God so loved the world, he didn't send a committee.

Still, there has to be a small knot of knowledgeable people who know the EE and software side of this. It needs to be small because it needs to be nimble, yet with enough technical depth to make things work. I've seen the agony of trying to define a standard (i.e., the X3J11 committee to write the first standard for the C language)...it is a formidable task, and the difficult increases with the size of the committee. There are all kinds of technical details to think about and the Atlas bus would at least be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE

On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <arvid.evans@...> wrote:

If hams were to design and standardize a ham-bus-system approach to modular equipment,

should the design be placed in public domain, or under one of the free-to-use licenses???

Should a group be formed specifically for the purpose of bus design, documentation, and
publication?

How would upgrades, modifications, and alternative bus designs be handled?

Seems there are lots of questions, lots of possible opinions, and lots of work to do.

Arv

#52617

I took part in the IETF committee on voice over IP (now RFC3261) and the Unicode standardization, it was getting teeth pulled out. That said, we have a few standards that already know of in the analog world. We have the BNC and the SMA connectors with 50 ohms interconnections. We have the 12, 5 and 3.3v supply. We don't have a control bus, but there is always the I2C. We don't have a standard software API but we do have hamlib, two or three SDR standards for sample formats.?

The good thing about standards, said Rciahrd Stevens, is that there are so many to choose from. That's the trouble!! Any attempt at protocol for radios will emerge as either cripplingly illogical like the CAT protocol or extensively overdone like the SIP protocol. What to include and what to leave out often results in such massively systems that implementing that in itself becomes a very challenging process, leave along hacking it.

The cubesat world has a physical standard, the PC104E boards. Each board is 90 mm by 95 mm with four mounting holes that are not symmetric at all (so smart, einstien!). The pinout is not defined either.?

In my lab, everything is BNC, everything is 50 ohms, everything has on-board regulators to bring things down from 12v to the required voltages. I have 10-12 of BNC pigtails, each about 6 inches. That, a pair of RF attenuators, two VOMs, a specan and the occasional Rigol DS1052E is my 'bus'. The trouble with physically standardization, it might not suit the RF layouts. For instance, an race ring mixer needs the LO, RF, IF at specific locations. They can't be moved. Instead, I have standardized on tack soldering BNC connectors exactly where I need them on the ground plane.?

- f?

Show quoted text

On Thu, Jun 21, 2018 at 8:20 AM, Jack Purdum via Groups.Io <jjpurdum@...> wrote:

Arv:

All tough questions, but worth answering. Think of the possibilities. Personally, I think this should be of the free-to-use license, mainly because I don't want to see extensions to the bus that aren't under someone's control. Otherwise, the standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I had on my desk when I was the department chairman:

??? For God so loved the world, he didn't send a committee.

Still, there has to be a small knot of knowledgeable people who know the EE and software side of this. It needs to be small because it needs to be nimble, yet with enough technical depth to make things work. I've seen the agony of trying to define a standard (i.e., the X3J11 committee to write the first standard for the C language)...it is a formidable task, and the difficult increases with the size of the committee. There are all kinds of technical details to think about and the Atlas bus would at least be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE

On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <arvid.evans@...> wrote:

If hams were to design and standardize a ham-bus-system approach to modular equipment,
should the design be placed in public domain, or under one of the free-to-use licenses???
Should a group be formed specifically for the purpose of bus design, documentation, and
publication?
How would upgrades, modifications, and alternative bus designs be handled?

Seems there are lots of questions, lots of possible opinions, and lots of work to do.

Arv

w7hd.rh

#52619

开云体育

I like the Atlas bus because it is intended for RF connections.? It has terminations, it has adequate pins for most potential applications, AND is off-the-shelf technology.? The test sweeps reveal a well thought out design.? Kudos to them! Ron W7HD

On 06/20/2018 07:50 PM, Jack Purdum via Groups.Io wrote:

Arv:

All tough questions, but worth answering. Think of the possibilities. Personally, I think this should be of the free-to-use license, mainly because I don't want to see extensions to the bus that aren't under someone's control. Otherwise, the standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I had on my desk when I was the department chairman:

??? For God so loved the world, he didn't send a committee.

Still, there has to be a small knot of knowledgeable people who know the EE and software side of this. It needs to be small because it needs to be nimble, yet with enough technical depth to make things work. I've seen the agony of trying to define a standard (i.e., the X3J11 committee to write the first standard for the C language)...it is a formidable task, and the difficult increases with the size of the committee. There are all kinds of technical details to think about and the Atlas bus would at least be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE

On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <arvid.evans@...> wrote:

If hams were to design and standardize a ham-bus-system approach to modular equipment,

should the design be placed in public domain, or under one of the free-to-use licenses???

Should a group be formed specifically for the purpose of bus design, documentation, and
publication?

How would upgrades, modifications, and alternative bus designs be handled?

Seems there are lots of questions, lots of possible opinions, and lots of work to do.

Arv

-- 
Ron W7HD - NAQCC#7587 OMISS#9898 KX3#6966 LinuxUser#415320
Editor OVARC newsletter

Rahul Srivastava

#52624

Ron ,

?Atlas bus has its own share of problems with the clk flowing thru bus. Secondly with the new gigabit protocol situation would become worse, decent synchronisation of Mercury cards was very tough. Signal integrity is much better on single board radios compared to the modular HPSDR unit.

73

Rahul VU3WJM?

#52637

An Elephant is a Mouse designed by that same committee.

Jack Purdum

#52644

Rahul:

Good to know. Failures are learning experiences, too.

Jack, W8TEE

On Wednesday, June 20, 2018, 11:54:37 PM EDT, Rahul Srivastava via Groups.Io <vu3wjm@...> wrote:

Ron ,

?Atlas bus has its own share of problems with the clk flowing thru bus. Secondly with the new gigabit protocol situation would become worse, decent synchronisation of Mercury cards was very tough. Signal integrity is much better on single board radios compared to the modular HPSDR unit.

73

Rahul VU3WJM?

w7hd.rh

#52645

开云体育

iT SURE LOOKED GOOD in the ad :-)? The basic idea is great, but I knew some tweaking/redesign would be in order.? Plus, it would be expensive to implement. Thanks for letting me know!!! Ron W7HD

On 06/20/2018 08:54 PM, Rahul Srivastava via Groups.Io wrote:

Ron ,

?Atlas bus has its own share of problems with the clk flowing thru bus. Secondly with the new gigabit protocol situation would become worse, decent synchronisation of Mercury cards was very tough. Signal integrity is much better on single board radios compared to the modular HPSDR unit.

73

Rahul VU3WJM?

-- 
Ron W7HD - NAQCC#7587 OMISS#9898 KX3#6966 LinuxUser#415320
Editor OVARC newsletter

[email protected]

#52654

I've worked with computer buses since the early '70s.? Most buses get standardized about the time they become obsolete.? Trying to set a standard simply forces people to look elsewhere for something better, which they also want to standardize.? And so it goes...
--
Jon Titus, KZ1G
Herriman, UT USA

#52663

On computers, a standard bus makes sense, they are just shipping around data and don't care
exactly how that happens so long as it is fast and error free.? And maybe some source of power.

Requirements of a "standard bus" for a radio are going to be all over the map, depending on?
what the radio does, frequencies of interest, how the boards are split up, power levels, ...

A radio with a motherboard into which various modules get plugged into is not a bad idea.
The PCIe connectors themselves work well, are cheap, mate with just a properly routed PC board.?
But that motherboard will be an ad-hoc collection of signals, nothing like implementing a PCIe
bus for a PC.? Unless, possibly, it's an SDR.??

Modules are good though.? I like the thought of a bunch of boards that plug together into a stack
on 0.1" headers, rather like Arduino shields.? And perhaps slides into one of the many extruded?
aluminum boxes with rails.? ?

On the uBitx, might be two boards split between the first mixer and the 30mhz LPF,
so the SSB exciter could be used with various versions of the PA, with its various filters and power?
levels and frequencies.? A third board could optionally be stacked on top of the PA for an
automatic antenna tuner for those keen to have something special to talk about on the air.
Different PA boards for those wanting cheap and functional vs those insisting on "real" RF parts.
Perhaps a different PA board with rx preamp and SWR meter for 1296 mhz.??

But for all those cases, the connection between boards could be kept quite simple.? Just one RF wire
plus power plus an i2c/spi bus.? ?I would not want the burden of some $100 motherboard that
attempts to "do everything" (whatever that is) when I join those boards.

Of course, you may someday want another optional board in the stack on the other side of
the uBitx exciter to implement digital modes using an RPi.? ?And a 7" display stacked on top of that.
Those interconnections might be quite different than that between the exciter and the PA.
But easily handled by a different header from exciter to the digital board.

Stacks of boards can be very tough to tinker with.? I prefer to keep connections mostly along one edge
of the board so a short flexible cable (or U-shaped assembly of two connectors and a small board)
can be slipped in there and the boards folded out for easy debug.? Ideally, use the far side of the board
for a connection to the next board up so you can fold out all three flat on the desk.?

Of course, I've never built a radio like that.??
And will likely have a totally different idea of how best to go about it once I do.
But a generic ham-bus that satisfies everybody seems unlikely.

Jerry, KE7ER?

?

Show quoted text

On Thu, Jun 21, 2018 at 07:40 am, <tituskz1g@...> wrote:

I've worked with computer buses since the early '70s.? Most buses get standardized about the time they become obsolete.? Trying to set a standard simply forces people to look elsewhere for something better, which they also want to standardize.? And so it goes...

#52664

Jack

After attempting to discuss this with a local ham, it became obvious that the idea

of a "Standard Ham Radio Bus" is many things to many people, and maybe not a

manageable concept.? It sort of fits with the old saying that "one should pick battles

that can be won".? Probably not even a committee could decide whether this bus

includes I2C/TWI, RF-signal, RF-power, virtual? links, radio links, 24V/12V/5V/3.3V,

keying, PTT, Linear control, regulated voltages, unregulated voltages, USB, Ethernet,

POE, CAT, terminations, etc.? And some have already suggested that it might be

focused particularly on support of modular BITX specific design blocks.? Maybe the

concept is best kept within the group members who are doing modular builds of BITX

and BITX related designs??

Arv
_._

Show quoted text

On Wed, Jun 20, 2018 at 8:50 PM Jack Purdum via Groups.Io <jjpurdum=[email protected]> wrote:

Arv:

All tough questions, but worth answering. Think of the possibilities. Personally, I think this should be of the free-to-use license, mainly because I don't want to see extensions to the bus that aren't under someone's control. Otherwise, the standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I had on my desk when I was the department chairman:

??? For God so loved the world, he didn't send a committee.

Still, there has to be a small knot of knowledgeable people who know the EE and software side of this. It needs to be small because it needs to be nimble, yet with enough technical depth to make things work. I've seen the agony of trying to define a standard (i.e., the X3J11 committee to write the first standard for the C language)...it is a formidable task, and the difficult increases with the size of the committee. There are all kinds of technical details to think about and the Atlas bus would at least be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE

On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <arvid.evans@...> wrote:

If hams were to design and standardize a ham-bus-system approach to modular equipment,
should the design be placed in public domain, or under one of the free-to-use licenses???
Should a group be formed specifically for the purpose of bus design, documentation, and
publication?
How would upgrades, modifications, and alternative bus designs be handled?

Seems there are lots of questions, lots of possible opinions, and lots of work to do.

Arv

#52667

I have been a ham for over 40 years and in all that time there is not even a 'standard' for the microphone jack.? Even if the plugs are the same type , the wiring is often different.? Even the 12 V dc connections are often different.? Seems that the pl259 and bnc are about all that has been mostly constant.

ku4pt

Show quoted text

On Thu, Jun 21, 2018 at 12:14 PM, Arv Evans <arvid.evans@...> wrote:

Jack

After attempting to discuss this with a local ham, it became obvious that the idea
of a "Standard Ham Radio Bus" is many things to many people, and maybe not a
manageable concept.? It sort of fits with the old saying that "one should pick battles
that can be won".? Probably not even a committee could decide whether this bus
includes I2C/TWI, RF-signal, RF-power, virtual? links, radio links, 24V/12V/5V/3.3V,
keying, PTT, Linear control, regulated voltages, unregulated voltages, USB, Ethernet,
POE, CAT, terminations, etc.? And some have already suggested that it might be
focused particularly on support of modular BITX specific design blocks.? Maybe the
concept is best kept within the group members who are doing modular builds of BITX
and BITX related designs??

Arv
_._

On Wed, Jun 20, 2018 at 8:50 PM Jack Purdum via Groups.Io <jjpurdum=[email protected]> wrote:
Arv:

All tough questions, but worth answering. Think of the possibilities. Personally, I think this should be of the free-to-use license, mainly because I don't want to see extensions to the bus that aren't under someone's control. Otherwise, the standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I had on my desk when I was the department chairman:

??? For God so loved the world, he didn't send a committee.

Still, there has to be a small knot of knowledgeable people who know the EE and software side of this. It needs to be small because it needs to be nimble, yet with enough technical depth to make things work. I've seen the agony of trying to define a standard (i.e., the X3J11 committee to write the first standard for the C language)...it is a formidable task, and the difficult increases with the size of the committee. There are all kinds of technical details to think about and the Atlas bus would at least be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE

On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <arvid.evans@...> wrote:

If hams were to design and standardize a ham-bus-system approach to modular equipment,
should the design be placed in public domain, or under one of the free-to-use licenses???
Should a group be formed specifically for the purpose of bus design, documentation, and
publication?
How would upgrades, modifications, and alternative bus designs be handled?

Seems there are lots of questions, lots of possible opinions, and lots of work to do.

Arv

Jack Purdum

#52668

Jerry:

All thoughtful points, many of which I didn't even consider. I remember trouble shooting the S100 bus using "extender cards". It made it possible, but it wasn't ideal. Any "rank mount" has to keep that in mind.

It seems to me that the bus bottleneck is most severe on the transmit side, whereas the receive side might lend itself to more standardization. Personally and for obvious reasons, I want to mess around with SDR rigs. I don't know enough to write one...yet, but I would like to learn about them. IT just makes sense to push things out of hardware and into software where possible. That said, I don't want to be tethered to a PC or laptop in the process. I want the computing horsepower to be part of the rig.

As I said, I don't have the talent to carry this forward, but segmenting the xmit/rec elements might help define a bus that could work.

Jack, W8TEE

On Thursday, June 21, 2018, 11:41:33 AM EDT, Jerry Gaffke via Groups.Io <jgaffke@...> wrote:

On computers, a standard bus makes sense, they are just shipping around data and don't care
exactly how that happens so long as it is fast and error free.? And maybe some source of power.

Requirements of a "standard bus" for a radio are going to be all over the map, depending on?
what the radio does, frequencies of interest, how the boards are split up, power levels, ...

A radio with a motherboard into which various modules get plugged into is not a bad idea.
The PCIe connectors themselves work well, are cheap, mate with just a properly routed PC board.?
But that motherboard will be an ad-hoc collection of signals, nothing like implementing a PCIe
bus for a PC.? Unless, possibly, it's an SDR.??

Modules are good though.? I like the thought of a bunch of boards that plug together into a stack
on 0.1" headers, rather like Arduino shields.? And perhaps slides into one of the many extruded?
aluminum boxes with rails.? ?

On the uBitx, might be two boards split between the first mixer and the 30mhz LPF,
so the SSB exciter could be used with various versions of the PA, with its various filters and power?
levels and frequencies.? A third board could optionally be stacked on top of the PA for an
automatic antenna tuner for those keen to have something special to talk about on the air.
Different PA boards for those wanting cheap and functional vs those insisting on "real" RF parts.
Perhaps a different PA board with rx preamp and SWR meter for 1296 mhz.??

But for all those cases, the connection between boards could be kept quite simple.? Just one RF wire
plus power plus an i2c/spi bus.? ?I would not want the burden of some $100 motherboard that
attempts to "do everything" (whatever that is) when I join those boards.

Of course, you may someday want another optional board in the stack on the other side of
the uBitx exciter to implement digital modes using an RPi.? ?And a 7" display stacked on top of that.
Those interconnections might be quite different than that between the exciter and the PA.
But easily handled by a different header from exciter to the digital board.

Stacks of boards can be very tough to tinker with.? I prefer to keep connections mostly along one edge
of the board so a short flexible cable (or U-shaped assembly of two connectors and a small board)
can be slipped in there and the boards folded out for easy debug.? Ideally, use the far side of the board
for a connection to the next board up so you can fold out all three flat on the desk.?

Of course, I've never built a radio like that.??
And will likely have a totally different idea of how best to go about it once I do.
But a generic ham-bus that satisfies everybody seems unlikely.

Jerry, KE7ER?

?

Show quoted text

On Thu, Jun 21, 2018 at 07:40 am, <tituskz1g@...> wrote:

I've worked with computer buses since the early '70s.? Most buses get standardized about the time they become obsolete.? Trying to set a standard simply forces people to look elsewhere for something better, which they also want to standardize.? And so it goes...

#52672

I think you'll find the so239/pl259 combination as well as the bnc were
first standardized by the military and then the commercial industry
picked them up.

Unless you can find a bus standard that has been adopted by the
military first you probably aren't going to find wide acceptance.

tim ab0wr

On Thu, 21 Jun 2018 12:22:54 -0400
"Ralph Mowery" <rmowery42@...> wrote:

I have been a ham for over 40 years and in all that time there is not
even a 'standard' for the microphone jack.  Even if the plugs are the
same type , the wiring is often different.  Even the 12 V dc
connections are often different.  Seems that the pl259 and bnc are
about all that has been mostly constant.

ku4pt


On Thu, Jun 21, 2018 at 12:14 PM, Arv Evans <arvid.evans@...>
wrote:

Jack

After attempting to discuss this with a local ham, it became
obvious that the idea
of a "Standard Ham Radio Bus" is many things to many people, and
maybe not a
manageable concept.  It sort of fits with the old saying that "one
should pick battles
that can be won".  Probably not even a committee could decide
whether this bus
includes I2C/TWI, RF-signal, RF-power, virtual  links, radio links,
24V/12V/5V/3.3V,
keying, PTT, Linear control, regulated voltages, unregulated
voltages, USB, Ethernet,
POE, CAT, terminations, etc.  And some have already suggested that
it might be
focused particularly on support of modular BITX specific design
blocks. Maybe the
concept is best kept within the group members who are doing modular
builds of BITX
and BITX related designs?

Arv
_._


On Wed, Jun 20, 2018 at 8:50 PM Jack Purdum via Groups.Io
<jjpurdum= [email protected]> wrote:  
 
Arv:

All tough questions, but worth answering. Think of the
possibilities. Personally, I think this should be of the
free-to-use license, mainly because I don't want to see extensions
to the bus that aren't under someone's control. Otherwise, the
standard slowly dissolves into chaos.

Every time an organization question comes up, I think of a sign I
had on my desk when I was the department chairman:

  *  For God so loved the world, he didn't send a committee.*

Still, there has to be a *small *knot of knowledgeable people who
know the EE and software side of this. It needs to be small
because it needs to be nimble, yet with enough technical depth to
make things work. I've seen the agony of trying to define a
standard (i.e., the X3J11 committee to write the first standard
for the C language)...it is a formidable task, and the difficult
increases with the size of the committee. There are all kinds of
technical details to think about and the Atlas bus would at least
be a thoughtful starting point.

I think this could be a rewarding endeavor.

Jack, W8TEE



On Wednesday, June 20, 2018, 8:36:53 PM EDT, Arv Evans <  
arvid.evans@...> wrote:  


If hams were to design and standardize a ham-bus-system approach to
modular equipment,
should the design be placed in public domain, or under one of the
free-to-use licenses?
Should a group be formed specifically for the purpose of bus
design, documentation, and
publication?
How would upgrades, modifications, and alternative bus designs be
handled?

Seems there are lots of questions, lots of possible opinions, and
lots of work to do.

Arv
 
 

#52674

A PA board up against an exciter is a recipe for trouble.
Potentially has 40 dBm signals very near some -20 dBm signals.
Digital stuff near the exciter could also cause trouble.
Could be dealt with by having the two ground planes face each other.
And/or, extra long headers between the boards and a sheet of metal
placed into the intervening extruded aluminum slot.
And/or the careful placement of parts, one board relative to the other.?
And/or shielding around sensitive circuits, perhaps 1 cm long sections of
copper plumbing pipe with a copper sheet (a PCB?) bolted over the top
of the entire board with appropriate RFI gasket material.

Show quoted text

On Thu, Jun 21, 2018 at 08:41 am, Jerry Gaffke wrote:

On the uBitx, might be two boards split between the first mixer and the 30mhz LPF,
so the SSB exciter could be used with various versions of the PA, with its various filters and power?
levels and frequencies.?

Jack Purdum

#52675

Arv:

My EE knowledge is like going to an axe fight with a squirt gun. Perhaps the best I can hope for realistically is a modular approach. Separating into function blocks like we see in equipment discussions might be the best I can hope for...and I can live with that. Farhan, for example, simplified modifying/replacing the LCD display by simply using the pin connections between the main board and the Raduino. A few more pins from the main board IF would have made the JackAl board interface virtually plug-and-play. Building with that kind of flexibility would be good enough for most of us who enjoy tinkering without having to resort to a "ham bus".

The good news is that I'm not smart enough to even join an EE battle, let alone pick one. I'll try to stand in the background, nodding occasionally so I look like I understand what's going on, and wait to the EE guys to model a workable module approach. That, too, would be fun.

Jack, W8TEE

On Thursday, June 21, 2018, 12:15:06 PM EDT, Arv Evans <arvid.evans@...> wrote: