K3: S-meter calibration (redux)

[Bill VanAlstyne W5WVO]

To recap: In a thread on this list back when the K3 was first announced, there was a good deal of debate about how many dB an S-unit should be on a calibrated S-meter. It was pointed out that if an S-unit is 6 dB, then a signal at the receiver noise floor would be between S1 and S2 on the meter, which seems rather high; a 5-dB S-unit, for example, would put the noise floor a lot closer to where it intuitively belongs on the meter. But it was also pointed out that the 6-dB S-unit has very deep historical roots, and in IARU Region 2, it's a published spec, not a matter of endless debate.

All this discussion was based on the universal assumption that S9 is by definition hard-pegged at 50 uV. It was the one assumption no one questioned.

Today, ARRL Labs' review of the new TenTec OMNI VII was published on their website (members only), and it was revealed that the OMNI VII S-meter is in fact accurately calibrated at 6 dB per S-unit -- but that S9 is pegged at 67 uV, not 50 uV! Using this higher threshold for S9 and a 6-dB S-unit, the noise floor drops down to where it should be -- somewhere between S0 and S1. A 10 dB S+N/N signal of 0.5 uV comes in around an intuitively reasonable S2.

To me, this seems like an elegant and creative solution to the "6-dB problem," drawing a sensible compromise between tradition and engineering common sense. It will be interesting to see where the K3 comes down in all this. There seems little doubt that the K3 and the OMNI VII (in that order, #1 and #2) are going to dominate the "center" of the transceiver market for a long time. By all rights, IKenSu should be completely shut out once both the K3 and the OMNI VIII are in the marketplace. They will have to depend on marketing hype and brand loyalty -- but who knows, that may be enough to keep them in the game until they can catch up. IF they can catch up.  :-)

Bill / W5WVO
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Ron D'Eau Claire-2]

You make a good argument, Bill.

It's important to note that the S-unit is a subjective "by-ear" evaluation
of a signal strength.

When someone manufactures a product, it's important to know how to determine
if it was assembled and working as expected. It was for those reasons that
various manufacturers came up with ways of adjusting the receiver to produce
certain readings on the "S-Meter" when test signals of certain levels were
applied to the input.

The only historically (and in practice correct) measurement is to listen to
the signal by ear and judge the "S-reading". If the meter doesn't agree with
what you decided by listening to the signal without referring to the meter,
the meter is wrong.

Ron AC7AC



-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Bill W5WVO
Sent: Wednesday, June 06, 2007 6:39 PM
To: [Elecraft]
Subject: [Elecraft] K3: S-meter calibration (redux)


To recap: In a thread on this list back when the K3 was first announced,
there was a good deal of debate about how many dB an S-unit should be on a
calibrated S-meter. It was pointed out that if an S-unit is 6 dB, then a
signal at the receiver noise floor would be between S1 and S2 on the meter,
which seems rather high; a 5-dB S-unit, for example, would put the noise
floor a lot closer to where it intuitively belongs on the meter. But it was
also pointed out that the 6-dB S-unit has very deep historical roots, and in
IARU Region 2, it's a published spec, not a matter of endless debate.

All this discussion was based on the universal assumption that S9 is by
definition hard-pegged at 50 uV. It was the one assumption no one
questioned.

Today, ARRL Labs' review of the new TenTec OMNI VII was published on their
website (members only), and it was revealed that the OMNI VII S-meter is in
fact accurately calibrated at 6 dB per S-unit -- but that S9 is pegged at 67
uV, not 50 uV! Using this higher threshold for S9 and a 6-dB S-unit, the
noise floor drops down to where it should be -- somewhere between S0 and S1.
A 10 dB S+N/N signal of 0.5 uV comes in around an intuitively reasonable S2.

To me, this seems like an elegant and creative solution to the "6-dB
problem," drawing a sensible compromise between tradition and engineering
common sense. It will be interesting to see where the K3 comes down in all
this. There seems little doubt that the K3 and the OMNI VII (in that order,
#1 and #2) are going to dominate the "center" of the transceiver market for
a long time. By all rights, IKenSu should be completely shut out once both
the K3 and the OMNI VIII are in the marketplace. They will have to depend on
marketing hype and brand loyalty -- but who knows, that may be enough to
keep them in the game until they can catch up. IF they can catch up.  :-)

Bill / W5WVO

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[wayne burdick]

For pragmatic reasons I prefer about 4 dB/S-unit, which is the default
for the K3. The operator can set it differently using the menu.

73,
Wayne
N6KR


---

http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill VanAlstyne W5WVO]

Ron D'Eau Claire wrote:

> The only historically (and in practice correct) measurement is to
> listen to the signal by ear and judge the "S-reading". If the meter
> doesn't agree with what you decided by listening to the signal
> without referring to the meter, the meter is wrong.

I agree, Ron, in the context of giving on-air signal reports to stations as
you establish QSO with them. I always give by-ear signal reports, which
basically means going by the old established verbal definitions -- Barely
detectable (S1), Weak (S3), Fair (S5), and so forth.

Having said that, a calibrated strength meter affords the capability to make
reasonably accurate strength comparisons between different signals when doing
so makes sense. But why use S-units on the meter at all? There was a very
reasonable argument made back during the original incarnation of this thread
(by Craig, VK3HE) that a calibrated strength meter shouldn't be an "S-meter"
at all, since the original subjective, relative meaning of a strength report
couldn't possibly be meaningful when rendered by an instrument making an
absolute, out-of-context measurement -- and that the strength meter really
should read out directly in dBuV. According to Craig, all modern commercial
and military communications receivers are now doing this.

I think this argument holds a lot of merit, but my guess is that many (if not
most) hams would howl in protest at not having an "S-meter" on their rig. So
if we have to have one of the blinkin' things, it should at least read out
something that is more or less sensible. Having the no-antenna receiver noise
floor reading out at S1-S2 is not sensible IMO, so if you must have a 6-dB
S-unit, the S9 signal level has to go higher than 50 uV. In short, I think
TenTec got it right.

Bill / W5WVO

>
> Ron AC7AC
>
>
>
> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]] On Behalf Of Bill W5WVO
> Sent: Wednesday, June 06, 2007 6:39 PM
> To: [Elecraft]
> Subject: [Elecraft] K3: S-meter calibration (redux)
>
>
> To recap: In a thread on this list back when the K3 was first
> announced, there was a good deal of debate about how many dB an
> S-unit should be on a calibrated S-meter. It was pointed out that if
> an S-unit is 6 dB, then a signal at the receiver noise floor would be
> between S1 and S2 on the meter, which seems rather high; a 5-dB
> S-unit, for example, would put the noise floor a lot closer to where
> it intuitively belongs on the meter. But it was also pointed out that
> the 6-dB S-unit has very deep historical roots, and in IARU Region 2,
> it's a published spec, not a matter of endless debate.
>
> All this discussion was based on the universal assumption that S9 is
> by definition hard-pegged at 50 uV. It was the one assumption no one
> questioned.
>
> Today, ARRL Labs' review of the new TenTec OMNI VII was published on
> their website (members only), and it was revealed that the OMNI VII
> S-meter is in fact accurately calibrated at 6 dB per S-unit -- but
> that S9 is pegged at 67 uV, not 50 uV! Using this higher threshold
> for S9 and a 6-dB S-unit, the noise floor drops down to where it
> should be -- somewhere between S0 and S1. A 10 dB S+N/N signal of 0.5
> uV comes in around an intuitively reasonable S2.
>
> To me, this seems like an elegant and creative solution to the "6-dB
> problem," drawing a sensible compromise between tradition and
> engineering common sense. It will be interesting to see where the K3
> comes down in all this. There seems little doubt that the K3 and the
> OMNI VII (in that order, #1 and #2) are going to dominate the
> "center" of the transceiver market for a long time. By all rights,
> IKenSu should be completely shut out once both the K3 and the OMNI
> VIII are in the marketplace. They will have to depend on marketing
> hype and brand loyalty -- but who knows, that may be enough to keep
> them in the game until they can catch up. IF they can catch up.  :-)
>
> Bill / W5WVO
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill VanAlstyne W5WVO]

Hi Wayne,

In addition to the slope of the function (dB per S-unit), is the scale set
point user-configurable? Or is this hard-coded such that S9=50 uV?

Bill / W5WVO


wayne burdick wrote:

> For pragmatic reasons I prefer about 4 dB/S-unit, which is the default
> for the K3. The operator can set it differently using the menu.
>
> 73,
> Wayne
> N6KR
>
>
> ---
>
> http://www.elecraft.com
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com 

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[wayne burdick]

On Jun 6, 2007, at 10:01 PM, Bill W5WVO wrote:

> Hi Wayne,
>
> In addition to the slope of the function (dB per S-unit), is the scale
> set point user-configurable? Or is this hard-coded such that S9=50 uV?

User settable.

Wayne


---

http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Bill VanAlstyne W5WVO]

You guys are awesome!  :D

Bill / /W5WVO


wayne burdick wrote:

> On Jun 6, 2007, at 10:01 PM, Bill W5WVO wrote:
>
>> Hi Wayne,
>>
>> In addition to the slope of the function (dB per S-unit), is the
>> scale set point user-configurable? Or is this hard-coded such that
>> S9=50 uV?
>
> User settable.
>
> Wayne
>
>
> ---
>
> http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[dj7mgq]

> You guys are awesome!  :D
>
>>> In addition to the slope of the function (dB per S-unit), is the
>>> scale set point user-configurable? Or is this hard-coded such that
>>> S9=50 uV?
>>
>> User settable.

I'm not sure I really like this at all. S-meters are meaningless if they
do not adhere to some kind of norm.

The IARU Region I published a "norm" many years ago and, imho, it is a
blemish on the face of the face of amateur community that this has not
been implemented on a wide scale basis.

It is beyond me how regulators can take interference reports based on
s-meter values seriously. The s-meter is for many amateurs around the
world, the only way they have of selectively measuring low level
signals. One can not expect an affordable s-meter to be accurate to a
tenth of a dB but plus/minus one or two dBs ought to be possible.

vy 73 de toby

PS: The IARU Region I recommendation:
Page 156 in http://www.iaru-r1.org/VHF_Handbook_V5_11.pdf
Page 142 in http://www.iaru-r1.org/HFM%20Handbook%20V6.pdf
--
DD5FZ, 4N6FZ (ex dj7mgq, dg5mgq, dd5fz)
K2 #885, K2/100 #3248
K3/100 #??? (< #200)
DOK C12, BCC, DL-QRP-AG
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Greg - AB7R]

The good new Toby is that for folks who are strictly concerned about a "norm" is that you
can calibrate it to whatever norm you wish.  I think that's the whole point here.  Not
everyone is so concerned about an exact norm nor does everyone use their S-meter when giving
a signal report....I sure don't....to me it's how I hear the signal on the radio.  If I need
to make a comparison between two or more signals then I might use the S-meter.

So don't fret...and adjust away to suit your needs.

73
Greg
AB7R


On Thu Jun  7  9:24 , Toby Deinhardt  sent:

>> You guys are awesome!  :D
>>
>>>> In addition to the slope of the function (dB per S-unit), is the
>>>> scale set point user-configurable? Or is this hard-coded such that
>>>> S9=50 uV?
>>>
>>> User settable.
>
>I'm not sure I really like this at all. S-meters are meaningless if they
>do not adhere to some kind of norm.
>
>The IARU Region I published a "norm" many years ago and, imho, it is a
>blemish on the face of the face of amateur community that this has not
>been implemented on a wide scale basis.
>
>It is beyond me how regulators can take interference reports based on
>s-meter values seriously. The s-meter is for many amateurs around the
>world, the only way they have of selectively measuring low level
>signals. One can not expect an affordable s-meter to be accurate to a
>tenth of a dB but plus/minus one or two dBs ought to be possible.
>
>vy 73 de toby
>
>PS: The IARU Region I recommendation:
>Page 156 in http://www.iaru-r1.org/VHF_Handbook_V5_11.pdf
>Page 142 in http://www.iaru-r1.org/HFM%20Handbook%20V6.pdf
>--
>DD5FZ, 4N6FZ (ex dj7mgq, dg5mgq, dd5fz)
>K2 #885, K2/100 #3248
>K3/100 #??? (
>DOK C12, BCC, DL-QRP-AG
>_______________________________________________
>Elecraft mailing list
>Post to: [hidden email]
>You must be a subscriber to post to the list.
>Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft   
>
>Help: http://mailman.qth.net/subscribers.htm
>Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

On Jun 7, 2007, at 6:24 AM, Toby Deinhardt wrote:

> It is beyond me how regulators can take interference reports based  
> on s-meter values seriously. The s-meter is for many amateurs  
> around the world, the only way they have of selectively measuring  
> low level signals. One can not expect an affordable s-meter to be  
> accurate to a tenth of a dB but plus/minus one or two dBs ought to  
> be possible.

I agree 100%.

I like the idea that by referencing a fixed standard, i.e. S9 = 50uV  
into 50ohms, and then with a fixed slope after that, i.e. 6dB/S-unit,  
I can get an actual calculation of path loss by knowing the rest of  
the kit, e.g. line loss, antenna gain, transmit power, etc. That is  
really useful! Randomly changing S-meter behavior because it "looks  
good" or "sounds good" seems pretty darned counter-productive to me.

As a minimum, if someone makes a 10dB change in their signal, I  
should see a 10dB change on my meter. I just roll my eyes when  
someone kicks on their amplifier and I see a 3 S-unit change. Oh please!

So this gets back to a discussion of user interface. Microsoft has  
convinced us that being able to change things is somehow useful and  
desirable when, in fact, all it really does is cause confusion and  
support problems. Almost nothing is more frustrating than finding  
that the person you are trying to help has "customized" their system  
beyond recognition and nothing you tell them is really going to help  
them get it to work.

Sorry Wayne, but being able to change the S-meter slope and intercept  
strikes me as being a bad option. To me that is like changing the  
calibration of a voltmeter or wattmeter because you like the needle  
pointer to move differently.

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[wayne burdick]

Brian Lloyd wrote:

> ....being able to change the S-meter slope and intercept strikes me as
> being a bad option. To me that is like changing the calibration of a
> voltmeter or wattmeter because you like the needle pointer to move
> differently.

I'd agree with you if there were a single world-wide standard for S-9,
and no need to compensate for slight differences in receive gain from
one unit to the next. But reality is that S-meters usually require both
scale and offset calibration.

This is also more flexible. As I mentioned earlier, I set my S-meters
up for 4 dB per S-unit. Here's why: I like a greater degree of
sensitivity in the S-meter so I can see the effects of things like
preamp on/off, filter changes, notch, NR, etc. It also makes band-pass
filters easier to tweak when there isn't a scope or AF voltmeter handy,
and you can more readily see the effect of an improved antenna during
A/B testing.

If hams wanted to be precise in assessing signal levels, we'd report
them in dBm and do a lot of averaging. But for most operators this is a
hobby, not a job  :)

73,
Wayne
N6KR

---

http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Leigh L. Klotz Jr WA5ZNU]

I wholly and completely disagree.
1. S Meter standardization is a failed effort.  S meters are "marketing"
numbers.  dBuV or dBmW is the measure to use.  If you are filing
interference reports with the FCC and cannot figure out how to convert
your signal strength readings to dB relative units, you need to make
better measurements.
2. If Elecraft chooses to allow end-users to calibrate and set their S
Meter readings to have a known intercept (50uV at S9) and slope (4dB,
6dB, 3dB), so much the better, as it helps number one.
3. And, although I would not hold MSFT responsible for pushing forward
UI design, I certainly don't think they offer the best options in terms
of user configurability!

Leigh/WA5ZNU
On Thu, 7 Jun 2007 9:36 am, Brian Lloyd wrote:

> On Jun 7, 2007, at 6:24 AM, Toby Deinhardt wrote:
>
>> It is beyond me how regulators can take interference reports based  on
>> s-meter values seriously. The s-meter is for many amateurs  around the
>> world, the only way they have of selectively measuring  low level
>> signals. One can not expect an affordable s-meter to be  accurate to a
>> tenth of a dB but plus/minus one or two dBs ought to  be possible.
>
> I agree 100%.
>
> I like the idea that by referencing a fixed standard, i.e. S9 = 50uV  
> into 50ohms, and then with a fixed slope after that, i.e. 6dB/S-unit,  
> I can get an actual calculation of path loss by knowing the rest of  
> the kit, e.g. line loss, antenna gain, transmit power, etc. That is  
> really useful! Randomly changing S-meter behavior because it "looks  
> good" or "sounds good" seems pretty darned counter-productive to me.
>
> As a minimum, if someone makes a 10dB change in their signal, I  should
> see a 10dB change on my meter. I just roll my eyes when  someone kicks
> on their amplifier and I see a 3 S-unit change. Oh please!
>
> So this gets back to a discussion of user interface. Microsoft has  
> convinced us that being able to change things is somehow useful and  
> desirable when, in fact, all it really does is cause confusion and  
> support problems. Almost nothing is more frustrating than finding  that
> the person you are trying to help has "customized" their system  beyond
> recognition and nothing you tell them is really going to help  them get
> it to work.
>
> Sorry Wayne, but being able to change the S-meter slope and intercept  
> strikes me as being a bad option. To me that is like changing the  
> calibration of a voltmeter or wattmeter because you like the needle  
> pointer to move differently.
>
> 73 de Brian, WB6RQN
> Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com
>
>
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft    Help:
> http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

On Jun 7, 2007, at 9:50 AM, wayne burdick wrote:

> Brian Lloyd wrote:
>
>> ....being able to change the S-meter slope and intercept strikes  
>> me as being a bad option. To me that is like changing the  
>> calibration of a voltmeter or wattmeter because you like the  
>> needle pointer to move differently.
>
> I'd agree with you if there were a single world-wide standard for  
> S-9, and no need to compensate for slight differences in receive  
> gain from one unit to the next. But reality is that S-meters  
> usually require both scale and offset calibration.

Yes, they do. But they can be calibrated to a standard regardless of  
the gain of the radio.

> This is also more flexible. As I mentioned earlier, I set my S-
> meters up for 4 dB per S-unit. Here's why: I like a greater degree  
> of sensitivity in the S-meter so I can see the effects of things  
> like preamp on/off, filter changes, notch, NR, etc. It also makes  
> band-pass filters easier to tweak when there isn't a scope or AF  
> voltmeter handy, and you can more readily see the effect of an  
> improved antenna during A/B testing.

What you are doing is changing the calibration instead of changing  
the *resolution*. What you really want to do is to be able to resolve  
smaller changes easily. So blow up the scale. Add calibration points  
for half S-units. That would give you 3dB points on the meter. That  
is even better than your 4dB resolution!

> If hams wanted to be precise in assessing signal levels, we'd  
> report them in dBm and do a lot of averaging.

I agree. I would prefer to have a meter calibrated in dBm but we have  
used S-units for so long that it is part of the fabric. Heck, we  
still use the English Standard system of measurements in the US. I  
have to switch back and forth between metric and ES all the time. And  
sometimes it is convenient to measure resistance in ohms or  
conductance in mohs even though we know they are really the same thing.

> But for most operators this is a hobby, not a job  :)

It is a technical hobby. We measure voltage, resistance, current, and  
power to very accurate levels. Why should we therefore say that  
accurately measuring receive signal level is unimportant? You  
yourself say that you use the S-meter to:

"...see the effects of things like preamp on/off, filter changes,  
notch, NR, etc. It also makes band-pass filters easier to tweak when  
there isn't a scope or AF voltmeter handy, and you can more readily  
see the effect of an improved antenna during A/B testing."

Clearly you are using it as an instrument of measurement. Why not  
have it conform to a standard so that the readings are useful rather  
than just randomly relative?

I think this gets back to my comment about resolution. If you are  
using a quantized bar-graph display it is easier to change the  
calibration than to change the resolution. OTOH, three digits would  
be nice or even an analog meter. (I actually still prefer analog  
meters for a lot of things, especially doing calibrations involving  
tweaking things.)

Never mind.

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Darwin, Keith]

I disagree with this.  If S-meter cal was indeed a standard, that's one
thing but the reality of S-meters is that they are all over the map.

If I calibrate my rig and know how many dB per s-unit than I'm far ahead
of my peers who only *think* they know.  If I have 4 dB per unit I
expect to see a 2.5 unit jump when the other guy kicks on the linear.

- Keith N1AS -
- K2 5411.ssb.100 -

-----Original Message-----

Sorry Wayne, but being able to change the S-meter slope and intercept
strikes me as being a bad option. To me that is like changing the
calibration of a voltmeter or wattmeter because you like the needle
pointer to move differently.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

On Jun 7, 2007, at 9:50 AM, Leigh L Klotz, Jr. wrote:

> I wholly and completely disagree.

Interesting. It seems to me that we are saying the same thing.

> 1. S Meter standardization is a failed effort.  S meters are  
> "marketing" numbers.  dBuV or dBmW is the measure to use.  If you  
> are filing interference reports with the FCC and cannot figure out  
> how to convert your signal strength readings to dB relative units,  
> you need to make better measurements.

Precisely.

> 2. If Elecraft chooses to allow end-users to calibrate and set  
> their S Meter readings to have a known intercept (50uV at S9) and  
> slope (4dB, 6dB, 3dB), so much the better, as it helps number one.

That Elecraft allows it to be set to a standard and that it will be  
consistent from day-to-day and band-to-band is fantastic. It means it  
CAN be used for measurement. The only question is what the  
calibration will be. Since we have been taught that one S-unit  
represents a 6dB change, new kit should adhere to that. When I sit  
down at your radio and see a 1 S-unit change, I should be able to  
know what that means without having to ask you, "Hey Leigh, what are  
your S-meter calibration coefficients?" I don't have to do that with  
a wattmeter or a voltmeter, do I?

> 3. And, although I would not hold MSFT responsible for pushing  
> forward UI design, I certainly don't think they offer the best  
> options in terms of user configurability!

I think Microsoft has their head up their ... uh ... well, they are  
pretty clueless when it comes to coming up with a good UI. Their  
options su... are supremely suboptimal.

So, seems you are agreeing with me and I with you.

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

On Jun 7, 2007, at 10:12 AM, Darwin, Keith wrote:

> I disagree with this.  If S-meter cal was indeed a standard, that's  
> one
> thing but the reality of S-meters is that they are all over the map.
>
> If I calibrate my rig and know how many dB per s-unit than I'm far  
> ahead
> of my peers who only *think* they know.  If I have 4 dB per unit I
> expect to see a 2.5 unit jump when the other guy kicks on the linear.

I find it interesting that people say they disagree with me and then  
go on to agree with me.

The points I am making are:

1. Most radios have S-meters that don't tell you anything useful  
other than, "the signal is bigger or smaller." Why bother with the  
meter if your ear can do just as well and probably more accurately?

2. Just because everyone else has crap for a meter doesn't mean you  
should too.

3. Having a meter that accurately tells you the received signal  
strength, one that can be accurately turned into relative dB changes  
or, better still, changes relative to 0dBm, is a very good thing.

4. Being able to sit down at a random radio, look at its meter, and  
know what it is telling you without having to get a lesson from its  
owner would be really nice.

The only question is what the calibration should be. If the FCC  
didn't put it on the test and if there weren't a european standard, I  
would say "hey, knock yourself out; do what makes you happy so long  
as it is consistent."

But we do actually have a standard and we do (finally) have a radio  
that can perform accurate measurement to that standard. Why throw  
that away?

Sorry, I have carried on too long about this. When I get my K3 I can  
set its meter to give me real measurements in S-units. If you use my  
radio you will know that what you had to learn for your FCC exam is  
actually reflected in the behavior of the meter in my radio.

Oh, and I want a meter with resolution that corresponds to the  
accuracy of the measurement. ;-)

Uh, and can I have it for $59 too? :-)

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[wayne burdick]

> I think this gets back to my comment about resolution. If you are
> using a quantized bar-graph display it is easier to change the
> calibration than to change the resolution. OTOH, three digits would be
> nice or even an analog meter. (I actually still prefer analog meters
> for a lot of things, especially doing calibrations involving tweaking
> things.)

I just added a task to my future-K3-firmware list: allow the VFO B
display to optionally show signal level in dBm. Three digits. Just for
you, Brian   :)

73,
Wayne
N6KR


---

http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Brian Lloyd-6]

On Jun 7, 2007, at 10:35 AM, wayne burdick wrote:

>> I think this gets back to my comment about resolution. If you are  
>> using a quantized bar-graph display it is easier to change the  
>> calibration than to change the resolution. OTOH, three digits  
>> would be nice or even an analog meter. (I actually still prefer  
>> analog meters for a lot of things, especially doing calibrations  
>> involving tweaking things.)
>
> I just added a task to my future-K3-firmware list: allow the VFO B  
> display to optionally show signal level in dBm. Three digits. Just  
> for you, Brian   :)

<sitting here with jaw hanging open> Wow. Uh, thanks!

73 de Brian, WB6RQN
Brian Lloyd - brian HYPHEN wb6rqn AT lloyd DOT com


_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[dj7mgq]

Hallo,

Wayne wrote:
> I'd agree with you if there were a single world-wide standard for S-9,
> and no need to compensate for slight differences in receive gain from

There is a standard in Region 1, but, to be honest, a standard which up
to now has almost totally been ignored by the industry. This has been
bemoaned in almost all serious rig tests I can remember here in Germany.
When a rig has a s-meter which comes close to the Region 1 standard, it
is often considered by the authors to be a real plus point for the rig.

Wayne wrote:
> This is also more flexible. As I mentioned earlier, I set my S-meters up
> for 4 dB per S-unit. Here's why: I like a greater degree of sensitivity

Sorry but I do not agree.

A signal strength meter, which is what an s-meter should be, should not
be affected by changes in amplification after entering the rig, i.e. the
reading should not be affected by a built-in preamplifier.

Wayne wrote:
> If hams wanted to be precise in assessing signal levels, we'd report
> them in dBm and do a lot of averaging. But for most operators this is a
> hobby, not a job  :)

I would prefer an s-meter which showed dBm or dBu and was fairly
accurate. For many, if not most amateurs, +/- 1.5dB would, I think, be
more than good enough. Professional style measurements are not really
what you would want to do with an amateur communications receiver, but
nonetheless the numbers from one rig to the next still ought to be
comparable.

BTW: To quote the head honcho of our local club: We do professional
amateur radio... ;-)

Leigh wrote:
 > numbers.  dBuV or dBmW is the measure to use.  If you are filing
 > interference reports with the FCC and cannot figure out how to convert

In some parts of the world and for some budgets I agree, in other cases,
all you might have is a s-meter. Then it would be more than just nice to
know that the numbers mean actually something, and that you can then say
+/- 3dB the following level in dBm can be seen at the antenna jack. Easy
math if the s-meter sticks to a standard.

Granted with the K3, you can go into menus, and set it up the way you
need it. However, this does lead to feel good metering, i.e. useless for
the other station, who does not know how you've set your rig up.


vy 73 de toby

PS: RST should be more than just "Raten, Schätzen, Täuschen"
--
DD5FZ, 4N6FZ (ex dj7mgq, dg5mgq, dd5fz)
K2 #885, K2/100 #3248
K3/100 #??? (< #200)
DOK C12, BCC, DL-QRP-AG
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[dj7mgq]

> I just added a task to my future-K3-firmware list: allow the VFO B
> display to optionally show signal level in dBm. Three digits. Just for
> you, Brian   :)

One can not help but love these guys.....

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com