Topics

New Direct Sampling Transceiver IC from Texas Instruments #hardware

erik@...
 

Sure it was a simplistic example, but I was comparing it an existing SDR, not tractors or livestock.

jdow
 

So do I9 processor chips.
{^_-}

On 20190305 22:59:09, Simon Brown wrote:
Also,
Look at the power dissipation, these needs a heatsink and cooling.
Simon Brown, G4ELI
www.sdr-radio.com
-----Original Message-----
From: main@airspy.groups.io <main@airspy.groups.io> On Behalf Of jdow
Sent: 06 March 2019 06:37
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware
Well, let's think about this. With a 3 GHz I/Q sample rate, if that is how the chip works out, you get a noise power of -174 dBm basic noise figure 1 noise plus about 95 dB leading to a base level for the A/D noise level of about -79 dBm. Without any decimation tricks that means the top end of the dynamic range is about 0 dBm. That is very respectable for any receiver. Put filters on that thing and you get very serious levels of dynamic range. Then use decimation to extend your LSBs downwards. You can increase the top end dB per dB with what decimation buys you. This toy could have a staggering dynamic range anywhere in a 1+ GHz frequency range 0 through 1+ GHs. This is the stuff that allows ad hoc repeaters without exotic filters, for example.
{^_^}

Ralph A. Schmid, dk5ras
 

Exactly, you got it!! This beast opens up an exciting era for the not so price sensitive :)

With his unit a nice SDR in a small 100*160 casing in the 5k range should be possible that make a nice heart for almost any RF application in normal consumer and commercial frequency bands.

Ralph.

-----Original Message-----
From: main@airspy.groups.io [mailto:main@airspy.groups.io] On Behalf Of
jdow
Sent: Wednesday, March 6, 2019 7:37 AM
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas
Instruments #hardware

Well, let's think about this. With a 3 GHz I/Q sample rate, if that is how the
chip works out, you get a noise power of -174 dBm basic noise figure 1 noise
plus about 95 dB leading to a base level for the A/D noise level of about -79
dBm. Without any decimation tricks that means the top end of the dynamic
range is about 0 dBm. That is very respectable for any receiver. Put filters on
that thing and you get very serious levels of dynamic range. Then use
decimation to extend your LSBs downwards. You can increase the top end dB
per dB with what decimation buys you. This toy could have a staggering
dynamic range anywhere in a 1+ GHz frequency range 0 through 1+ GHs. This
is the stuff that allows ad hoc repeaters without exotic filters, for example.

{^_^}

On 20190305 22:09:44, erik@... wrote:
Downloaded the spec sheet - looks wonderful, but the price man.

Surprisingly 14bit too, I *think* hams are looking to 16 bit
resolution for the future are they not? Especially for this price bracket.

By comparison, Ettus B210 uses AD9361 - which I find today at $238
USD, whilst a completed B210 SDR sells for $1,216 USD (and that is
friendly USA prices), *so a
x10 markup* (but understandable).

The AFE7444
<https://c212.net/c/link/?t=0&l=en&o=2390916-
1&h=2410926770&u=http%3A%
2F%2Fwww.ti.com%2Fproduct%2FAFE7444%2Fsamplebuy%3FHQS-asc-dc-
hsc-afe74
xx-pr-sa-afe7444-wwe&a=AFE7444> I see online at retailers for $1802
USD. So I guess you have to know how to roll your own, or expect a
device nearly $15 - $20k USD.

At least it wouldn't launch with a USB interface :)

In any regard, at least we know what to expect in hobbyland, for the
next 5 - 10 years in SDR terms...

Peace.

Ralph A. Schmid, dk5ras
 

Direct sampling makes the price; look into it, no LO :)

 

Ralph.

 

From: main@airspy.groups.io [mailto:main@airspy.groups.io] On Behalf Of erik@...
Sent: Wednesday, March 6, 2019 7:10 AM
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware

 

Downloaded the spec sheet - looks wonderful, but the price man.

Surprisingly 14bit too, I *think* hams are looking to 16 bit resolution for the future are they not? Especially for this price bracket.

By comparison, Ettus B210 uses AD9361 - which I find today at $238 USD, whilst a completed B210 SDR sells for $1,216 USD (and that is friendly USA prices), so a x10 markup (but understandable).

The AFE7444 I see online at retailers for $1802 USD. So I guess you have to know how to roll your own, or expect a device nearly $15 - $20k USD.

At least it wouldn't launch with a USB interface :)

In any regard, at least we know what to expect in hobbyland, for the next 5 - 10 years in SDR terms...

Peace.

Ralph A. Schmid, dk5ras
 

Exactly. When I saw the specs, I expected a high 3 digit price – now that it even got low 4digit came not as the biggest surprise.

 

Also the specs of the additional hardware and software to make use of those specs will be above the normal hobbyists equipment.

 

Ralph, dk5ras.

 

From: main@airspy.groups.io [mailto:main@airspy.groups.io] On Behalf Of August Johnson
Sent: Tuesday, March 5, 2019 5:14 PM
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware

 

That line means pricing EACH if you buy 100 units. Not price for 100 units. You will be disappointed. $1800 in single unit quantities. Click on the "order now" link and TRY to add ONE to your cart. You'll see $1801.70.

 

August KG7BZ

 

 

On Tue, Mar 5, 2019 at 8:03 AM Siegfried Jackstien <siegfried.jackstien@...> wrote:

so ... when do start a group order???

should be easy o find a cheap chinese pcb manufacturer hi hi

(just hoping a bit)

dg9bfc sigi

ps ... yes that is also what i have seen price per 100 units (half a metre cut from the reel or so)

 

 

 

 

 

p chinese pc

Am 05.03.2019 um 15:45 schrieb Fred Hillhouse:

Good eye David!

 

Product,Package, Pricing in 100-unit quantities, Order now at the TI store

AFE7444 quad-channel, wideband RF-sampling transceiver,  17-mm-by-17-mm, flip-chip ball grid array (FCBGA), Starting at US$1,749.90, AFE7444IABJ

AFE7422 dual-channel, wideband RF-sampling transceiver, 17-mm-by-17-mm FCBGA, Starting at US$1,249.90,  AFE7422IABJ

 

They are $17.50/piece and $12.50/piece which is not unreasonable.

 

There are evaluation modules available at $2499 and $1999.

 

 

 

-----Original Message-----
From: main@airspy.groups.io [mailto:main@airspy.groups.io] On Behalf Of David J Taylor via Groups.Io
Sent: Tuesday, March 05, 2019 9:36 AM
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware

 

From: Pierre Martel

 

at 1800$ a pop. Not sure it will be available in a near futur for our type

of use.

============================================

 

... and I read it as $1800 for a reel of 100!  $18 each.

 

Cheers,

David

--

SatSignal Software - Quality software for you

Web: http://www.satsignal.eu

Email: david-taylor@...

Twitter: @gm8arv

 

 

 

Simon Brown
 

Also,

Look at the power dissipation, these needs a heatsink and cooling.

Simon Brown, G4ELI
www.sdr-radio.com

-----Original Message-----
From: main@airspy.groups.io <main@airspy.groups.io> On Behalf Of jdow
Sent: 06 March 2019 06:37
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware

Well, let's think about this. With a 3 GHz I/Q sample rate, if that is how the chip works out, you get a noise power of -174 dBm basic noise figure 1 noise plus about 95 dB leading to a base level for the A/D noise level of about -79 dBm. Without any decimation tricks that means the top end of the dynamic range is about 0 dBm. That is very respectable for any receiver. Put filters on that thing and you get very serious levels of dynamic range. Then use decimation to extend your LSBs downwards. You can increase the top end dB per dB with what decimation buys you. This toy could have a staggering dynamic range anywhere in a 1+ GHz frequency range 0 through 1+ GHs. This is the stuff that allows ad hoc repeaters without exotic filters, for example.

{^_^}

jdow
 

Well, let's think about this. With a 3 GHz I/Q sample rate, if that is how the chip works out, you get a noise power of -174 dBm basic noise figure 1 noise plus about 95 dB leading to a base level for the A/D noise level of about -79 dBm. Without any decimation tricks that means the top end of the dynamic range is about 0 dBm. That is very respectable for any receiver. Put filters on that thing and you get very serious levels of dynamic range. Then use decimation to extend your LSBs downwards. You can increase the top end dB per dB with what decimation buys you. This toy could have a staggering dynamic range anywhere in a 1+ GHz frequency range 0 through 1+ GHs. This is the stuff that allows ad hoc repeaters without exotic filters, for example.

{^_^}

On 20190305 22:09:44, erik@... wrote:
Downloaded the spec sheet - looks wonderful, but the price man.
Surprisingly 14bit too, I *think* hams are looking to 16 bit resolution for the future are they not? Especially for this price bracket.
By comparison, Ettus B210 uses AD9361 - which I find today at $238 USD, whilst a completed B210 SDR sells for $1,216 USD (and that is friendly USA prices), *so a x10 markup* (but understandable).
The AFE7444 <https://c212.net/c/link/?t=0&l=en&o=2390916-1&h=2410926770&u=http%3A%2F%2Fwww.ti.com%2Fproduct%2FAFE7444%2Fsamplebuy%3FHQS-asc-dc-hsc-afe74xx-pr-sa-afe7444-wwe&a=AFE7444> I see online at retailers for $1802 USD. So I guess you have to know how to roll your own, or expect a device nearly $15 - $20k USD.
At least it wouldn't launch with a USB interface :)
In any regard, at least we know what to expect in hobbyland, for the next 5 - 10 years in SDR terms...
Peace.

Dana Myers
 

On 3/5/2019 10:09 PM, erik@... wrote:
Downloaded the spec sheet - looks wonderful, but the price man.

Surprisingly 14bit too, I *think* hams are looking to 16 bit resolution for the future are they not? Especially for this price bracket.

By comparison, Ettus B210 uses AD9361 - which I find today at $238 USD, whilst a completed B210 SDR sells for $1,216 USD (and that is friendly USA prices), so a x10 markup (but understandable).

Wouldn't you agree that's a bit of an apples and oranges comparison between
the AFE7444 direct-sampling at 3Gs/s with internal decimation and the B210
sampling at 62Ms/s with an RF down-converting front-end? ;-)

The AFE7444 I see online at retailers for $1802 USD. So I guess you have to know how to roll your own, or expect a device nearly $15 - $20k USD.

At least it wouldn't launch with a USB interface :)

In any regard, at least we know what to expect in hobbyland, for the next 5 - 10 years in SDR terms...

Peace.

erik@...
 

Downloaded the spec sheet - looks wonderful, but the price man.

Surprisingly 14bit too, I *think* hams are looking to 16 bit resolution for the future are they not? Especially for this price bracket.

By comparison, Ettus B210 uses AD9361 - which I find today at $238 USD, whilst a completed B210 SDR sells for $1,216 USD (and that is friendly USA prices), so a x10 markup (but understandable).

The AFE7444 I see online at retailers for $1802 USD. So I guess you have to know how to roll your own, or expect a device nearly $15 - $20k USD.

At least it wouldn't launch with a USB interface :)

In any regard, at least we know what to expect in hobbyland, for the next 5 - 10 years in SDR terms...

Peace.

doug
 


On 03/05/2019 04:23 PM, Marcus D. Leech wrote:
On 03/05/2019 04:18 PM, prog wrote:
If only one of these big companies (ADI, TI, Xilinx) could sponsor the development and build of a small batch for enthusiasts (who are mostly engineers anyway), it would accelerate the adoption of their products.

_.
As a retired RF Engineer, I have to say that new IC and transistor products are sampled to working engineers very quickly. As to adoption, it depends on the potential sales for any given product.
One company, which shall remain nameless, sampled me some very nice RF devices, I designed then into a product, and then I found that because we were not going to make many thousands of
the product, that they were not going to make the part any more. I was not the only engineer at the company who discovered a similar situation, and we no longer did business with this very major
supplier of RF semiconductors. Some reading this probably know the supplier I'm writing about.

As far as enthusiasts, it is obvious that there will be no production follow-on from the retired engineer or ham radio enthusiast working in his basement, so it would not be in the financial interest of
the supplier to make a development device available to these folks.

Some retired engineers are consultants, and it is possible that some companies will provide them with sample devices, but not development devices, but only those that are already in production.
I have done just a tiny bit of consulting, and I have received one or two device samples, so it does work out to that extent.

--doug, WA2SAY

Siegfried Jackstien
 

come on ....  it was a nice dream ... not?!?

now dream to add such a chip to pcb ... with a good fpga single board computer etc ...

if you are fast in developing you could earn a lot :-)

greetz

sigi dg9bfc

Am 05.03.2019 um 22:13 schrieb jdow:

No - if you buy in 100 unit quantities the price is $1800 each.
{^_^}

On 20190305 06:44:13, Siegfried Jackstien wrote:
100 ... for 1700 dollar ... that is 17 bucks per chip

now we make a group order hi hi

greetz

sigi dg9bfc

Am 05.03.2019 um 12:59 schrieb Ken Alexander:
Sounds cool, but at those prices you can count me out. 😖

Ken Alexander
So Phisai, Thailand
Blog: bueng-ken.com <http://bueng-ken.com>


On Tue, Mar 5, 2019 at 7:36 PM prog <@prog <mailto:@prog>> wrote:

    ADC: 3 GSPS
    DAC: 9 GSPS
    2 or 4 TRX channels

    Check this out!

https://news.ti.com/industrys-first-integrated-quad-and-dual-channel-rf-sampling-transceivers-enable-multiantenna-wideband-systems?HQS=asc-dcs-hsc-afe74xx-twit-lp-pr-wwe

jdow
 

No - if you buy in 100 unit quantities the price is $1800 each.
{^_^}

On 20190305 06:44:13, Siegfried Jackstien wrote:
100 ... for 1700 dollar ... that is 17 bucks per chip
now we make a group order hi hi
greetz
sigi dg9bfc
Am 05.03.2019 um 12:59 schrieb Ken Alexander:
Sounds cool, but at those prices you can count me out. 😖

Ken Alexander
So Phisai, Thailand
Blog: bueng-ken.com <http://bueng-ken.com>


On Tue, Mar 5, 2019 at 7:36 PM prog <@prog <mailto:@prog>> wrote:

ADC: 3 GSPS
DAC: 9 GSPS
2 or 4 TRX channels

Check this out!

https://news.ti.com/industrys-first-integrated-quad-and-dual-channel-rf-sampling-transceivers-enable-multiantenna-wideband-systems?HQS=asc-dcs-hsc-afe74xx-twit-lp-pr-wwe

jdow
 

Dream on!
{^_-}

On 20190305 06:36:10, David J Taylor via Groups.Io wrote:
From: Pierre Martel
at 1800$ a pop. Not sure it will be available in a near futur for our type of use.
============================================
... and I read it as $1800 for a reel of 100!  $18 each.
Cheers,
David

Marcus D. Leech
 

On 03/05/2019 04:18 PM, prog wrote:
If only one of these big companies (ADI, TI, Xilinx) could sponsor the development and build of a small batch for enthusiasts (who are mostly engineers anyway), it would accelerate the adoption of their products.

prog
 

If only one of these big companies (ADI, TI, Xilinx) could sponsor the development and build of a small batch for enthusiasts (who are mostly engineers anyway), it would accelerate the adoption of their products.

Marcus D. Leech
 

On 03/05/2019 03:04 PM, Ray WB6TPU wrote:
The Xilinx RFSOC parts also carry a price tag in the multi-K $ range as well. And they definitely are regulated by EAR/ITAR regulations.

-Ray WB6TPU
Full disclosure: I work at Xilinx.
I think the per-channel costs are a bit lower...



-----Original Message-----
From: main@airspy.groups.io [mailto:main@airspy.groups.io] On Behalf Of Marcus D. Leech via Groups.Io
Sent: Tuesday, March 05, 2019 10:53 AM
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware

EXTERNAL EMAIL

On 03/05/2019 01:14 PM, Martin Smith via Groups.Io wrote:
It is a fantastic chip, but ITAR would make it a pain to use in anything other than military applications ?
https://www.bis.doc.gov/ ->“Regulations”-> “EAR (Export Administration Regulations)”->“Category 3 - Electronics Design Development and Production”

Or am I missing something ?
There are these, as well:

https://www.xilinx.com/products/silicon-devices/soc/rfsoc.html

ADCs, FPGA, multi-core ARM all on the same chip...






 

The Xilinx RFSOC parts also carry a price tag in the multi-K $ range as well. And they definitely are regulated by EAR/ITAR regulations.

-Ray WB6TPU
Full disclosure: I work at Xilinx.

-----Original Message-----
From: main@airspy.groups.io [mailto:main@airspy.groups.io] On Behalf Of Marcus D. Leech via Groups.Io
Sent: Tuesday, March 05, 2019 10:53 AM
To: main@airspy.groups.io
Subject: Re: [airspy] New Direct Sampling Transceiver IC from Texas Instruments #hardware

EXTERNAL EMAIL

On 03/05/2019 01:14 PM, Martin Smith via Groups.Io wrote:
It is a fantastic chip, but ITAR would make it a pain to use in anything other than military applications ?
https://www.bis.doc.gov/ ->“Regulations”-> “EAR (Export Administration Regulations)”->“Category 3 - Electronics Design Development and Production”

Or am I missing something ?
There are these, as well:

https://www.xilinx.com/products/silicon-devices/soc/rfsoc.html

ADCs, FPGA, multi-core ARM all on the same chip...


Dana Myers
 

On 3/5/2019 10:14 AM, Martin Smith via Groups.Io wrote:
It is a fantastic chip, but ITAR would make it a pain to use in anything other than military applications ?
https://www.bis.doc.gov/ ->“Regulations”-> “EAR (Export Administration Regulations)”->“Category 3 - Electronics Design Development and Production”

Or am I missing something ?
I don't think you're missing a thing. The target markets for this
device are pretty clearly not consumer-oriented; I'd think
telecommunications infrastructure and terrestrial military
are more like it; I would be surprised if there's not a high-rel
version of this family for aero/space applications.

You can find the high-rel/rad-hard product folders on the
TI site; often, they are based on a "commercial" part like this.
It's always funny to click "Buy Now" and get the redirect to call
your local TI sales office.

Cheers,
Dana  K6JQ

Marcus D. Leech
 

On 03/05/2019 01:14 PM, Martin Smith via Groups.Io wrote:
It is a fantastic chip, but ITAR would make it a pain to use in anything other than military applications ?
https://www.bis.doc.gov/ ->“Regulations”-> “EAR (Export Administration Regulations)”->“Category 3 - Electronics Design Development and Production”

Or am I missing something ?
There are these, as well:

https://www.xilinx.com/products/silicon-devices/soc/rfsoc.html

ADCs, FPGA, multi-core ARM all on the same chip...


Marcus D. Leech
 

On 03/05/2019 01:14 PM, Martin Smith via Groups.Io wrote:
It is a fantastic chip, but ITAR would make it a pain to use in anything other than military applications ?
https://www.bis.doc.gov/ ->“Regulations”-> “EAR (Export Administration Regulations)”->“Category 3 - Electronics Design Development and Production”

Or am I missing something ?
I've put in a request for a full datasheet. That required that I satisfy ITAR muster.