[TenTec] ROB SHERWOOD's Comments on Transceiver Delay Times
rick@dj0ip.de
Rick at DJ0IP.de
Fri Dec 11 03:52:38 EST 2015
I forwarded our discussion to Rob.
Rob is not a member of this group, though he is in our Eagle and OM7 groups.
I did not cut off the long tail of previous emails, in case someone joins
this thread now after seeing Rob's name.
If anyone replies, please cut off all but Rob's emai. TNX.
Note: in his last paragraph, Rob is misquoting me - just as I misquoted him
earlier. I'm pushing Ten-Tec for an adjustable TXD with the high end being
25ms. It is currently 15ms on the OR2 and OM7 and 17ms on the Eagle. They
need to increase to at least 22ms for use with amplifiers with open-frame
relays.
TNX to Curt for relaying this thread to Rob, since my email to Rob got lost.
73 - Rick, DJ0IP
(Nr. Frankfurt, Germany)
===================================================================
From: Rob Sherwood [mailto:rob at sherweng.com]
Sent: Friday, December 11, 2015 7:19 AM
To: Rick - DJ0IP / NJ0IP
Subject: FW: Re: [TenTec] TenTec Digest, Vol 156, Issue 4
Here is the other post I made that went nowhere. Rob
===================================================================
From: Rob Sherwood
Sent: Monday, December 07, 2015 2:43 PM
To: mailto:tentec at contesting.com
Subject: RE: Re: [TenTec] TenTec Digest, Vol 156, Issue 4
Hi Rick,
I didn’t see your forward, so it may have gone to my other e-mail address.
Coming into the middle of a conversation may be a bit disjointed, so please
realize I don’t know what all has been said.
Having not made any latency measurements on the ANAN-200D yet, I am relying
on what NR0V told me. The keying was moved to hardware, and this solved the
problem. I have blind copied Warren on this reply, so possibly he can
comment directly. With a lot of things on my plate lately, including
several R-4Cs coming in for updating, plus a JRC NRD-515, I haven’t gotten
to several measurement projects.
I also spent a lot of time dealing with the incorrect transmitted phase
noise numbers published by the ARRL in the November QST for the K3S or K3
with the new synthesizer. The League has re-run their measurements and
corrected the Product Review that is npw On Line. The K3S is much better by
my measurements, and also the new measurements by the League, than
originally published. There was also a lot of time spent on some other K3S
measurements that I cannot go into right now.
Latency of the Flex 6000 series is on the order of 175 milliseconds. Having
just spent 17 hours operating the ARRL 160-meter CW contest this past
weekend, using a legacy (low latency) radio (TS-990S) and Alpha 89 PIN diode
QSK setup, I have some concerns about latency when it is over 50
milliseconds. I know that acoustically in a room 60 feet = 60 milliseconds,
and that kind of delay is an issue if someone is using a hearing assist
receiver, plus sound from speakers 60 feet away. The Kenwood has PIN diode
T/R switching like the Alpha. It is totally silent and very fast. I would
assume around 10 milliseconds with everything included. The 990S has
optional delays available in a menu, but with my linear that is completely
unnecessary. (The same would apply to an Alpha 87A.)
I believe K9CT ran Flex 6000 series radios this past weekend, or at least
they did in a recent contest. What happens if a 6000 is talking to another
6000? 175 milliseconds times 2 is quite a delay if one is running QSK.
Likely it is of less significance on SSB. Are there any RTTY experts on the
reflector that could comment about latency on that mode? Are there quick
breaks or back and forth transmissions on RTTY?
There are two issues here, latency and delay required for slow mechanical
T/R switches, such as old-fashioned open-frame relays such as in my old
Alpha 76PA. Never owned an L-4B or L-7, but they have the same issue. How
quickly an analog radio can switch between receiver and transmit is usually
measured by the League in product reviews. A look at the 6700 review in the
April 2015 issue of QST shows a T/R turn-around time of 198 milliseconds,
and an R/T turn-around time of 140 milliseconds. The measurement is
specified in SSB mode, as opposed to CW, for whatever reason.
Let’s take the average of those two measurements (169 milliseconds) times
two if two 6700s were talking to each other, and that is 338 milliseconds.
That would seem to be an issue when operating QSK. I know this weekend
there were times that I hit the F4 function key to transmit my call (N1MM),
and heard the station I was calling start transmitting again. I was able to
hit the escape key and cancel the rest of my transmission. I am guessing
that I was able to react within 50 milliseconds.
I think Rick has been pushing for some of the Ten-Tec rigs to have at least
17 milliseconds T/R delay to eliminate hot switching, I assume of the
linear. I honestly don’t know anything about the T/R switching in an Eagle
or an Omni-VII, whether it is solid-state or some kind of mechanical relay.
My old Icom IC-781 has some kind of tiny relay that must switch relatively
fast, and it does QSK just fine with an Alpha 99 which uses a vacuum relay
for T/R switching. That said, I prefer the PIN diode switching in the
TS-990S / Alpha 89 combo over the other pair.
That got kind of long! 73, Rob, NC0B
From: Curt Gamble [mailto:W0ALC at MINDSPRING.COM]
Sent: Monday, December 07, 2015 11:23 AM
To: Rob Sherwood
Subject: Fwd: Re: [TenTec] TenTec Digest, Vol 156, Issue 4
Rob,
Here is your chance!!!! SMILE!!!
-------- Forwarded Message --------
Subject:
Re: [TenTec] TenTec Digest, Vol 156, Issue 4
Date:
Mon, 7 Dec 2015 15:12:24 +0100
From:
rick at dj0ip.de <Rick at DJ0IP.de>
Reply-To:
Discussion of Ten-Tec Equipment <tentec at contesting.com>
To:
'Discussion of Ten-Tec Equipment' <tentec at contesting.com>
I copied Rob on my last message and hope that he enters the conversation.
I haven't measured any of this; I'm just quoting what Rob has told us.
On the ability to send good CW in the presence of latency, it also depends
on what is generating the tone that you are listening to. When we measured
this back in the 1970s, we were using a Drake C line and actually listening
to the transmitted signal. When we built our time sequencers for the
external keyer so that the L4B could be keyed before the T4XC, we played
with variable latency and that's when we noticed that 25mS was pretty much
the border or rather limit where the latency begin to bother the OP.
Of course if the sidetone generated by the radio is in real time, or less
than 25mS latency, then latency of the radio probably won't bother the op.
In our case, we were listening to a signal that was delayed by the amount of
TXD we inserted. This was in the late 70s so I'm struggling to remember
exact details, but Thomas (DL7AV) wrote it up in our magazine and I still
have the article - of course it's in German.
73 - Rick, DJ0IP
(Nr. Frankfurt, Germany)
-----Original Message-----
From: TenTec [mailto:tentec-bounces at contesting.com] On Behalf Of Barry N1EU
Sent: Monday, December 07, 2015 2:59 PM
To: Discussion of Ten-Tec Equipment
Subject: Re: [TenTec] TenTec Digest, Vol 156, Issue 4
I'm not sure that's the right question to ask because recovery doesn't
necessarily happen in real time. The Flex T/R cw recovery is really a mind
blower for an SDR. But does it necessarily mean you can do actual QSK and
hear someone else trying to break you in real time? I think there's still
receive latency so that even with the receiver recovering, it still lags
real time signals by 150msec (or whatever).
Barry N1EU
On Mon, Dec 7, 2015 at 8:50 AM, rick at dj0ip.de <Rick at dj0ip.de> wrote:
> Paul, when you release the PTT and the green line drops, how long does
> it take for the receiver to recover?
>
> 73 - Rick, DJ0IP
> (Nr. Frankfurt, Germany)
>
>
>
> -----Original Message-----
> From: TenTec [mailto:tentec-bounces at contesting.com] On Behalf Of Paul
> Christensen
> Sent: Monday, December 07, 2015 2:38 PM
> To: 'Discussion of Ten-Tec Equipment'
> Subject: Re: [TenTec] TenTec Digest, Vol 156, Issue 4
>
> > Sorry Barry, latency measured on the Anan does not necessarily apply
> > to
> the FLEX 6000.
>
> I made timing and latency measurements of my Flex 6700 using an
> Agilent 4-channel DSO. I captured the data and uploaded the DSO graphic
images.
> Keep an eye on the PTT trace from the external keyer as the timing
> reference. QSK performance has improved further since the date of these
> measurements. QSK performance is at least as good as the K3 and the many
> Ten Tecs I've owned. It is fast. At 40+ wpm, I don't feel like I'm
> dragging characters behind me due to any latency, and there are no
> headphone audio thumps or clicks in CW T/R transitions.
>
>
> https://community.flexradio.com/flexradio/topics/flex_6700_cw_keying_v
> _elecr
> aft_k3
>
> Paul, W9AC
>
> _______________________________________________
> TenTec mailing list
> TenTec at contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec
>
> _______________________________________________
> TenTec mailing list
> TenTec at contesting.com
> http://lists.contesting.com/mailman/listinfo/tentec
>
_______________________________________________
TenTec mailing list
TenTec at contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
_______________________________________________
TenTec mailing list
TenTec at contesting.com
http://lists.contesting.com/mailman/listinfo/tentec
________________________________________
If this email is spam, report it to www.OnlyMyEmail.com
More information about the TenTec
mailing list