No blank ESD diode position here?

1,029 views
Skip to first unread message

Jago Pearce

unread,
Oct 4, 2012, 9:38:59 AM10/4/12
to ultra-c...@googlegroups.com

I ordered a BAV99 ESD Diode assuming I would need one but when I opened the cover I found this. Do I need a diode because I can see no blank position?

http://s48.beta.photobucket.com/user/jago25_98/media/forums/bav99_zpse33377fe.jpg.html
http://i48.photobucket.com/albums/f211/jago25_98/forums/bav99_zpse33377fe.jpg

ik7tad op. Riccardo

unread,
Oct 4, 2012, 11:44:24 AM10/4/12
to ultra-c...@googlegroups.com
2012/10/4 Jago Pearce <jag...@gmail.com>:
>
> I ordered a BAV99 ESD Diode assuming I would need one but when I opened the
> cover I found this. Do I need a diode because I can see no blank position?

Perhaps you could watch the other side of the pcb too...

Chris Stake

unread,
Oct 4, 2012, 1:28:05 PM10/4/12
to ultra-c...@googlegroups.com

It’s hard to see what’s alongside C33, does it say U1? That’s exactly where the protection device(s) would be. Maybe they are already fitted?

Chris

 

 

--
You received this message because you are subscribed to the Google Groups "Ultra Cheap SDR" group.
To view this discussion on the web visit https://groups.google.com/d/msg/ultra-cheap-sdr/-/NiDbxii_iMYJ.
To post to this group, send email to ultra-c...@googlegroups.com.
To unsubscribe from this group, send email to ultra-cheap-s...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/ultra-cheap-sdr?hl=en.

Colorado Rob

unread,
Oct 4, 2012, 2:08:36 PM10/4/12
to ultra-c...@googlegroups.com

U1 is missing.  That's the same dongle I have.

darky

unread,
Oct 5, 2012, 4:24:57 AM10/5/12
to ultra-c...@googlegroups.com
The right position is U1, next to C33. On picture it appear to be empty.
 
BAV99 has a capacitance of 1.5pF, at 1GHz this is appr. 100 Ohm. You should be aware that this ESD diode will decrease your max op. frequency. This is the reason why they don't install them recently.
 
The right choice of diode would be a device with parasitic capacitance < 0.5 pF, special UHF ESD diode. Like :
 
 
 
Please, let me know do you understand the topic. Did I helped with this?

jdow

unread,
Oct 5, 2012, 4:37:24 AM10/5/12
to ultra-c...@googlegroups.com
Considering the loss of a meter of RG-174 coax at 1GHz the diode is not
that much of a killer. Your dongle is already rapidly becoming numb
without a preamp at the antenna.

{^_^}
> <http://s48.beta.photobucket.com/user/jago25_98/media/forums/bav99_zpse33377fe.jpg.html>
> http://i48.photobucket.com/albums/f211/jago25_98/forums/bav99_zpse33377fe.jpg <http://i48.photobucket.com/albums/f211/jago25_98/forums/bav99_zpse33377fe.jpg>

Gaston Picard

unread,
Oct 5, 2012, 9:39:24 AM10/5/12
to ultra-c...@googlegroups.com
That diodes look great!
Most of the usual diodes (not specifically designed for ESD control) are very bad at low voltages BAT54 ~ 10 pF at 0V, BAV99 looks better with its 1.5 pF...
I'd rather risk a small loss by inserting a BAV99, that probably is lower than what we have with the standard issue coaxial cable, than risk zapping the LNA input.

Gastón

Jago Pearce

unread,
Oct 7, 2012, 4:02:59 PM10/7/12
to ultra-c...@googlegroups.com
Ah thank you, I see it now. U1 is blank next to C33 so I need to fit a
protection diode.

The BAV99 diodes I have though have 3 legs and that position only has
2 points (not holes) for pins and they are very very small... smaller
than the BAV99 legs as yo ucan see.

Rob, were you able to fit a protection diode somehow?

pic for reference:
http://i48.photobucket.com/albums/f211/jago25_98/bav99_tooBig_zps2f743d50.jpg
--
Sites supported: http://moneybutnofixedabode.blogspot.com/
http://justfortheloveofit.org/

Colorado Rob

unread,
Oct 7, 2012, 10:13:12 PM10/7/12
to ultra-c...@googlegroups.com
No -- it's too late for me.  We'll see what the 820T dongle looks like when it arrives.  If it is missing as well, I will opt for the Infineon ESD protection diodes someone linked to earlier.

http://www.mouser.com/ProductDetail/Infineon/ESD0P2RF-02LRH-E6327/?qs=wK%252boHS4yu571cQ8LWymYuw%3d%3d

Colorado Rob

unread,
Oct 9, 2012, 8:32:33 PM10/9/12
to ultra-c...@googlegroups.com

New one arrived and it's fully protected. No blown LNA sounds great! :-)

limaunion

unread,
Oct 9, 2012, 10:39:21 PM10/9/12
to ultra-c...@googlegroups.com
good for you! where did you buy this esd protected dongle? mind to share the link? TIA

Colorado Rob

unread,
Oct 10, 2012, 12:06:30 AM10/10/12
to ultra-c...@googlegroups.com

Jago Pearce

unread,
Oct 10, 2012, 7:29:16 AM10/10/12
to ultra-c...@googlegroups.com
I don't understand why the Infineon ESD has 2 connections and my
BAV99 has 3... ?

How can I use this BAV99?

Lasse Radio

unread,
Oct 10, 2012, 7:41:58 AM10/10/12
to ultra-c...@googlegroups.com
These are the two types of dongles I have bought from Ebay (see image links)


The TV301 cost twice as much as the smaller DVB-T.  I'm currently waiting for my new ESD protected solder station, and when that arrives I'll solder a BAV99 on the DVB-T. If that works ok, I'll buy more of this cheap one.

Regards

Lasse

Colorado Rob

unread,
Oct 10, 2012, 8:33:02 AM10/10/12
to ultra-c...@googlegroups.com

My experience was the opposite. The expensive one lacked the protection diodes.

Both came in the same product packaging.  The shape and markings on the dongle are the same. The internal boards are different.

--
You received this message because you are subscribed to the Google Groups "Ultra Cheap SDR" group.
To view this discussion on the web visit https://groups.google.com/d/msg/ultra-cheap-sdr/-/czM4fWl_RLQJ.

jdow

unread,
Oct 10, 2012, 8:35:44 AM10/10/12
to ultra-c...@googlegroups.com
The BAV99 is more flexible than the Infineon part. Both form back to back
diodes. The single tab is the same as either single tab on the Infineon
part. The pair of tabs in parallel are equivalent to the other tab on
the Infineon part.

{^_^}

On 2012/10/10 04:29, Jago Pearce wrote:
> I don't understand why the Infineon ESD has 2 connections and my
> BAV99 has 3... ?
>
> How can I use this BAV99?
>
> On Wed, Oct 10, 2012 at 1:32 AM, Colorado Rob <colora...@gmail.com> wrote:
>> New one arrived and it's fully protected. No blown LNA sounds great! :-)
>>
>> On Oct 7, 2012 9:13 PM, "Colorado Rob" <colora...@gmail.com> wrote:
>>>
>>> No -- it's too late for me. We'll see what the 820T dongle looks like
>>> when it arrives. If it is missing as well, I will opt for the Infineon ESD
>>> protection diodes someone linked to earlier.
>>>
>>>
>>> http://www.mouser.com/ProductDetail/Infineon/ESD0P2RF-02LRH-E6327/?qs=wK%252boHS4yu571cQ8LWymYuw%3d%3d
>>>
>>>
>>> On Sun, Oct 7, 2012 at 3:02 PM, Jago Pearce <jag...@gmail.com> wrote:
>>>>
>>>> Ah thank you, I see it now. U1 is blank next to C33 so I need to fit a
>>>> protection diode.
>>>>
>>>> The BAV99 diodes I have though have 3 legs and that position only has
>>>> 2 points (not holes) for pins and they are very very small... smaller
>>>> than the BAV99 legs as yo ucan see.
>>>>
>>>> Rob, were you able to fit a protection diode somehow?
>>>>
>>>> pic for reference:
>>>>
>>>> http://i48.photobucket.com/albums/f211/jago25_98/bav99_tooBig_zps2f743d50.jpg
>>>>
>>>> On Thu, Oct 4, 2012 at 7:08 PM, Colorado Rob <colora...@gmail.com>
>>>> wrote:
>>>>> U1 is missing. That's the same dongle I have.
>>>>>
>>>>> On Oct 4, 2012 12:28 PM, "Chris Stake" <st...@btinternet.com> wrote:
>>>>>>
>>>>>> It�s hard to see what�s alongside C33, does it say U1? That�s exactly

Lasse Radio

unread,
Oct 10, 2012, 9:01:06 AM10/10/12
to ultra-c...@googlegroups.com
After I bought the cheaper DVB-T (without BAV99) I was especially looking for a dongle that confrmed had a ESD protection. It also was mentioned in the description text. So I had no worries about that.  I won't buy any dongle without a confirmed ESD protection, unless I can solder one myself without any risk. A dongle without it should only half or it's a rip off...

Regards

Lasse

limaunion

unread,
Oct 10, 2012, 9:07:59 PM10/10/12
to ultra-c...@googlegroups.com
unfortunately that seller doesn't ship to my country, i'm trying to find another one.

Gaston Picard

unread,
Oct 11, 2012, 7:40:34 AM10/11/12
to ultra-c...@googlegroups.com

Actually, it's not that way. BAV99 with both tabs on the "long side" soldered together is a couple of anti-parallel diodes. They limit voltage between the two equivalent electrodes to to +/- Vd, what should be about +/- 0.7V. The infineon part, on the other side, it has an equivalent of back-to-back diode and limits depending on the specific part. Like saying that is a "bidirectional zener", and the specific one:


is designed to operate at ~5V with minimum losses, and triggers at 7V.
IMO, using this part will probably let us with a fried LNA anyway as this chips are pretty sensitive. I wouldn't let anything higher than the specified maximum voltage (3.6V for the R820T) near the RF inputs... Most probably Infineon's is not the right kind of part, as both the EL4000 and R820T have a working voltage of 0V and will fry with more than 3.6V. I am going the anti-parallel BAV99 way.

Gaston

jdow

unread,
Oct 11, 2012, 8:33:32 AM10/11/12
to ultra-c...@googlegroups.com
The difference from what I described is that the Infineon part is not
a classic diode. It's a fairly fast 7 volt transient catcher. That makes
it good for some parts that can withstand 5 volts RMS without harm. For
those parts it would be good from an linearity standpoint.

The way E4000s seem to die around here I'm not sure they'd be happy at
even one volt RMS. I agree with you about the E4000 delicacy assessment.
Everybody (including infineon) is calling them diodes. So I'd presumed
they were diodes in the same sense as those in the BAV99 package with
the difference being the pinouts. I wonder how they managed to speed
up avalanche diodes enough to make their protection practical.

In either case the circuit hookup is as I described. And that was what
I was primarily addressing.

I think I might look into that Infineon device as protection for the
front end circuits in my semiconductor radios. (My R390A would shake
off even a 100V transient. So I'll simply rely on the spark gaps I
have when I use it.)

{^_-}
> --
> You received this message because you are subscribed to the Google Groups "Ultra
> Cheap SDR" group.
> To view this discussion on the web visit
> https://groups.google.com/d/msg/ultra-cheap-sdr/-/FfZPKnHeFAgJ.

Lasse Radio

unread,
Oct 11, 2012, 12:31:34 PM10/11/12
to ultra-c...@googlegroups.com


On Thursday, October 11, 2012 2:33:34 PM UTC+2, jdow wrote:
 (My R390A would shake off even a 100V transient. So I'll simply rely on the spark gaps I
have when I use it.)

Good Lord, that receiver is as old as I am... and probably in better shape.. ;-)

Regards

Lasse 

darky

unread,
Oct 11, 2012, 4:48:49 PM10/11/12
to ultra-c...@googlegroups.com

There are different protection schemes, all intended to clamp the input voltage to a certain value, it is easy to understand why diodes are used - because they clamp at the voltage when they start to conduct current.

Possible schematics :

1)

GND - diode - INPUT - diode - VDD - effectively clamping all voltages below (GND - Vdiode) and above (VDD + Vdiode)

Not always applicable because not all inputs can tolerate input voltage as high as VDD. Some special inputs work with voltages much lower then VDD.

2)

GND - diode - INPUT - effectively clamping all voltages below (GND - Vdiode), if the diode is a kind of zener diode then also clamping all voltages higher then Vzener

Zeners are usually with high parasitic capacitance and slow operating, not usable for RF, not usable as ESD protection. There are special diodes called TVS which are suitable for both :

www.semtech.com/images/datasheet/what_are_tvs_diodes.pdf

3)

GND - diode - diode - INPUT - usually the diodes are a kind of zener, i.e. TVS diodes or other possibility is that they form a diac, i.e. thyristor without gate

The idea is if the voltage on the output goes higher then Vth the diode configuration conductive effectively clamping the voltage to ground. Low capacitance ESD diodes are also suitable for RF. Have the advantage that the diodes are in series lowering the capacitance.

4)

GND - diode || diode - INPUT - two diodes in configuration back to back, effectively clamping any voltage lower or higher then Vdiode

Usable with general purpose diodes, Schottky diodes, special ESD diodes like Infineon or Semtech. For RF you definitely need special low capacitance diodes like TVS or Infineon's. Having the disadvantage that two diodes are in parallel, parasitic capacitance is in parallel.

Now let's go back to your question : "How to connect the ESD diode?"

Look at your PCB, what is your footprint? 2 pin, 3 pin, 4 pin ?

In case it is 2 pin you need a device like this :

http://www.infineon.com/dgdl/esd5v3l1u-02lrh.pdf?folderId=db3a30431f848401011fcbf2ab4c04c4&fileId=db3a3043284aacd8012895a71c230078

In case it is 3 pins :

http://www.infineon.com/dgdl/esd1p0rfseries.pdf?folderId=db3a30431441fb5d01149221c9230fb2&fileId=db3a30431441fb5d01149228afda0fc7

in case it is 4 pins :

http://www.infineon.com/dgdl/esd0p4rfl.pdf?folderId=db3a30431441fb5d01149221c9230fb2&fileId=db3a304318f3fe29011914ad9b81031f

Rule of tumb - choose a device with parasitic capacitance as low as possible. Example : 1 pF diode will have the 3 dB attenuation ( half the signal ) at about 3 GHz or when it is 50 Ohm. Calculator is available here :

http://pa2ohh.com/jslcimp.htm

Regards

jdow

unread,
Oct 12, 2012, 3:03:41 AM10/12/12
to ultra-c...@googlegroups.com
Unfortunately not. It needs a good dose of TLC and a new VFO to make it
nice. Then it needs some of the other tweaks I planned to make it REALLY
nice. They'll probably never be done, though. It's somewhat younger than
I am. I'm WW-II vintage myself.

{^_^}

darky

unread,
Oct 12, 2012, 4:55:12 AM10/12/12
to ultra-c...@googlegroups.com
Small clarification to my previous post :
 
Devices I posted in my previous post I choose only based on package and I did not look on the parameters, because I wanted to give an example, not to recommend a part.
 
If I need to look into the parameters I would choose this one :
 
 
or
 
 
I managed to see one of them here :
 
 
Good luck!
 
I just received my new sticks from eBay, they are R820T based and have the protection in place :
 

Dave Bell

unread,
Oct 13, 2012, 12:52:22 PM10/13/12
to ultra-c...@googlegroups.com

Where are you located? Seller states “Cannot be shipped to US or Canada”.

I’m going to email him and query whether that’s just covering himself against buyers who complain that it doesn’t work for TV…

Dave

Luis Franco

unread,
Oct 14, 2012, 1:59:02 PM10/14/12
to ultra-c...@googlegroups.com

Hi,

 

For those who have doubts about where BAV99 is connected, were are some pictures:

Picture 1: side by side of a E4K and a R820T.

Picture 2: E4000 with no protection diode.

Picture 3: E4000 with BAV99 protection diode.

 

Regards

 

Luis

To view this discussion on the web visit https://groups.google.com/d/msg/ultra-cheap-sdr/-/pGXBtQBK87MJ.


To post to this group, send email to ultra-c...@googlegroups.com.
To unsubscribe from this group, send email to ultra-cheap-s...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/ultra-cheap-sdr?hl=en.

 

__________ Informação do ESET Smart Security, versão da vacina 7567 (20121010) __________

 

A mensagem foi verificada pelo ESET Smart Security.

 

http://www.eset.com

 

__________ Informação do ESET Smart Security, versão da vacina 7569 (20121010) __________

 

A mensagem foi verificada pelo ESET Smart Security.

 

http://www.eset.com

 

__________ Informação do ESET Smart Security, versão da vacina 7580 (20121013) __________

 

A mensagem foi verificada pelo ESET Smart Security.

 

http://www.eset.com



__________ Informação do ESET Smart Security, versão da vacina 7580 (20121013) __________

A mensagem foi verificada pelo ESET Smart Security.

http://www.eset.com
1.JPG
2.JPG
3.JPG

Colorado Rob

unread,
Oct 14, 2012, 3:12:44 PM10/14/12
to ultra-c...@googlegroups.com
Interesting.  Thanks for posting this. I have the same pair of dongles.

Rob
WX9O

Luis Franco

unread,
Oct 14, 2012, 4:10:28 PM10/14/12
to ultra-c...@googlegroups.com

It´s always good to share experiences.

The E4000 dongle was modified by me. I used a small 16W soldering iron, and I have done it in 5 minutes. Just be careful with electrostatic electricity.

I did not notice differences in the dongle performance.

Now I can connect the e4000 to the external antenna, and safely compare the E4k and R820T.

 

With internal antenna, R820 seems to have more selectivity and gain.

 

Br

Luis

CT1EMY

ik7tad op. Riccardo

unread,
Oct 15, 2012, 12:28:11 PM10/15/12
to ultra-c...@googlegroups.com
2012/10/14 Luis Franco <lmvf...@gmail.com>:
>
> With internal antenna, R820 seems to have more selectivity and gain.
>


Hi Luis,
first of all, thanks for sharing your experience.
Then, what about frequency coverage of the two dongles?
I've heard that R820T can even tune down to 25 MHz, can you confirm it?
And what's the gain at those frequencies, particularly at 10m and 6 m?

73s de ik7tad, Riccardo.

Luis Franco

unread,
Oct 16, 2012, 6:06:32 PM10/16/12
to ultra-c...@googlegroups.com
Hi all,

Today I had same time to test and compare the E4000, R820T dongles and a
Icom IC-7000 ham transceiver.
I tested I amateur beacon on 144 Mhz, 70 Mhz and 50 MHz, with low signal.
The R820T works much better with more selectivity, but less gain than the
E4000.
I prefer the R820T, because for example the E4000 at 70 MHz it as lot of
interference from local FM ( 88-108 Mhz) radio stations.
The R820T sensitivity decreases below 70 MHz, but signal level low signals
are still audible.
I did not find any fix signal on 28 MHz, good for test.
When compared the R820T with a amateur radio transceiver, its works very
well. Not so good as the very expensive Icom transceiver, but quite good for
the money.
If I have time, I will post same videos :)

BR
Luis

-----Mensagem original-----
De: ultra-c...@googlegroups.com
[mailto:ultra-c...@googlegroups.com] Em nome de ik7tad op. Riccardo
Enviada: segunda-feira, 15 de Outubro de 2012 17:28
Para: ultra-c...@googlegroups.com
Assunto: Re: [ultra-cheap-sdr] No blank ESD diode position here?

--
You received this message because you are subscribed to the Google Groups
"Ultra Cheap SDR" group.
To post to this group, send email to ultra-c...@googlegroups.com.
To unsubscribe from this group, send email to
ultra-cheap-s...@googlegroups.com.
For more options, visit this group at
http://groups.google.com/group/ultra-cheap-sdr?hl=en.


__________ Informação do ESET Smart Security, versão da vacina 7580
(20121013) __________

A mensagem foi verificada pelo ESET Smart Security.

http://www.eset.com


__________ Informação do ESET Smart Security, versão da vacina 7588
(20121015) __________

A mensagem foi verificada pelo ESET Smart Security.

http://www.eset.com


__________ Informação do ESET Smart Security, versão da vacina 7592
(20121016) __________

A mensagem foi verificada pelo ESET Smart Security.

http://www.eset.com


__________ Informação do ESET Smart Security, versão da vacina 7593
(20121016) __________

A mensagem foi verificada pelo ESET Smart Security.

http://www.eset.com



__________ Informação do ESET Smart Security, versão da vacina 7593
(20121016) __________

Priyasloka Arya

unread,
Nov 14, 2012, 3:50:26 AM11/14/12
to ultra-c...@googlegroups.com
Hi
I am wondering what is this chip(marked in the photo). Few say latest SDR has LNA, is it a LNA chip.
Thanks
Priyasloka Arya

what This chip.jpg

Adam Nielsen

unread,
Nov 14, 2012, 4:40:15 AM11/14/12
to ultra-c...@googlegroups.com
> I am wondering what is this chip(marked in the photo). Few say latest SDR has
> LNA, is it a LNA chip.

You'll have to type out what it's got written on it. It's not possible to
identify just from the shape.

Cheers,
Adam.

ekow

unread,
Nov 14, 2012, 5:19:17 AM11/14/12
to ultra-c...@googlegroups.com

Let me guess
It's eeprom chip

Peter Hardy

unread,
Dec 26, 2012, 6:56:21 AM12/26/12
to ultra-c...@googlegroups.com
Hello i had problem with this two dvb-t tuners 1.LV5T deluxe,2.Hama nano 5330.
But i had external antena and 50 ohms cable.One day my dongle had small signal
can you help me where is ESD diode in this dougles.Thank you for help.exaple photo.

Dňa štvrtok, 4. októbra 2012 15:38:59 UTC+2 Jago Pearce napísal(-a):
LV5T deluxe up side.jpg
LV5T deluxe down side.jpg
Hama nano 5330 up side.jpg
Hama nano 5330 down side.jpg

Opendous Support

unread,
Dec 27, 2012, 12:30:20 AM12/27/12
to ultra-c...@googlegroups.com
The HamaNano seems to have a BAV99. It is the correct package and
location. Does the marking on the part say "A7"?
http://www.digikey.com/product-detail/en/BAV99LT1G/BAV99LT1GOSCT-ND

The FSC USB DVB-T appears to have an unpopulated 0402 footprint you
could solder ESD protection into.
http://www.digikey.com/product-detail/en/VCH4AG100R8MATWA/478-4709-1-ND

The LV5T has only one component which looks like it could be ESD
protection but its location and orientation imply it isn't.

In general, ESD protection is a component that connects the RF line
to GND before any other circuitry. It then shunts any high voltages
into ground before they can damage other components. DVB_ESD.jpg is a
picture of an EzTV dongle where I replaced the MCX connector with an
SMA connector and added a minimal loss attenuator to convert 50-Ohm to
75-Ohm. You can use this same method to add ESD protection to any
dongle at the RF connector. The AVX 0603 AntennaGuard is a good fit
if you can handle 0603 parts. A SOT23 BAV99 can be added this way,
with the two pins to GND and the single pin side to RF. For the very
paranoid you could also add a GDT. Or, add all 3 since they have
minimal insertion loss up to around 2GHz.
http://www.digikey.com/product-detail/en/VC06AG183R0YAT1A/478-5819-1-ND
http://www.digikey.com/product-detail/en/BAV99LT1G/BAV99LT1GOSCT-ND
http://www.digikey.com/product-detail/en/GTCS23-750M-R01-2/GTCS23-750M-R01-2CT-ND

If you are interested about the minimal loss attenuator, here are
some notes on impedance converters. However, mismatch losses are low
so you may as well ignore impedance mismatch of any adaptor you use.
http://www.ham-radio.com/k6sti/match.htm
HamaNano_ESD.jpg
FSC_USB_DVB-T_ESD.jpg
LV5T_ESD.jpg
DVB_ESD.jpg

David J Taylor

unread,
Dec 27, 2012, 2:56:49 AM12/27/12
to ultra-c...@googlegroups.com
-----Original Message-----
From: Opendous Support
[]
If you are interested about the minimal loss attenuator, here are
some notes on impedance converters. However, mismatch losses are low
so you may as well ignore impedance mismatch of any adaptor you use.
http://www.ham-radio.com/k6sti/match.htm
==================================================

Agreed. The so-called minimum loss resistive pad, with 5.6 dB loss, is
likely to be doing much more harm to the receiver sensitivity than simply
accepting the mismatch, as would be using either 50-ohm or 75-ohm cable in
place of the "correct" value.

For TV applications, these sticks may well be designed for a 75-ohm
connection (and possibly the chips will be so-designed as well, they may
just be measured in a 50-ohm environment for convenience), and at the higher
frequencies the lower loss of 75-ohm cable may outweigh the extra loss
through impedance mismatch. Has anyone seen the full specifications for
these chips? What input impedance provides best sensitivity, and how does
it vary with frequency? For all three chip types.

I'm not saying just ignore getting everything correct, as that will make for
the optimum system, simply that it's a grey area and worth doing the sums to
see just what the effects of mismatch and cable loss over long runs actually
is. Different for transmission, of course.

73,
David GM8ARV
--
SatSignal Software - Quality software written to your requirements
Web: http://www.satsignal.eu
Email: david-...@blueyonder.co.uk

Opendous Support

unread,
Dec 29, 2012, 2:53:25 PM12/29/12
to ultra-c...@googlegroups.com
>Has anyone seen the full specifications for these chips?

Your best bet if you want to get a sense of tuner specifications is
to look at the datasheets for similar products that aren't NDA'ed.
They don't mention anything but the nominal input impedance.
http://www.maximintegrated.com/datasheet/index.mvp/id/5508
http://www.maximintegrated.com/datasheet/index.mvp/id/5306
http://www.maximintegrated.com/datasheet/index.mvp/id/7260

>effects of mismatch and cable loss over long runs

A short impedance discontinuity should have negligible insertion
loss (S21) but significantly worse return loss (S11) according to the
attached QUCS simulations. I tested 50-ohm and 75-ohm BNC Jack-Jack
adapters with a VNWA. It looks like average insertion loss is lower
with the 75-ohm discontinuity probably due to lower losses of the
75-ohm adaptor (there is less dielectric material). Average VSWR
(S11) is worse, as expected. In any case, more proof you shouldn't
worry about mismatched connectors for receivers.
http://www.digikey.com/product-detail/en/CP-AD-555/367-1015-ND
http://www.digikey.com/product-detail/en/6737/501-1147-ND
ImpedanceDiscontinuityTesting-50ohm_Thru_Simulation.jpg
ImpedanceDiscontinuityTesting-75ohm_Thru_Simulation.jpg
ImpedanceDiscontinuityTesting-BNC_50ohm_Thru.jpg
ImpedanceDiscontinuityTesting-BNC_75ohm_Thru.jpg
ImpedanceDiscontinuity.sch
ImpedanceDiscontinuity.dpl

jdow

unread,
Dec 29, 2012, 9:08:55 PM12/29/12
to ultra-c...@googlegroups.com
In general the best noise figure you can do if you match the
signal source impedance in a receiver is 3 dB. The low noise
figure trick is to find the optimum mismatch. So there is no
particular advantage to being overly obsessive about matching
the 50 ohm impedance that isn't there. Of course, the closer
you get to having a source impedance that is optimal for the
receiver's noise figure the better it will be. But that's
usually when you are fighting tenths of a dB so you can meet
a customer's sometimes absurd noise figure demands. (More
than sometimes if the customer is peace time military. Oy!)

{^_^} Joanne, W6MKU
> ImpedanceDiscontinuityTesting-50ohm_Thru_Simulation.jpg
>
>
>
> ImpedanceDiscontinuityTesting-75ohm_Thru_Simulation.jpg
>
>
>
> ImpedanceDiscontinuityTesting-BNC_50ohm_Thru.jpg
>
>
>
> ImpedanceDiscontinuityTesting-BNC_75ohm_Thru.jpg
>
>
Reply all
Reply to author
Forward
0 new messages