Imput impedance of common emitter amplifier at RF frequencie

[bitrex]

On 7/31/19 5:40 PM, bitrex wrote:

On 7/31/19 1:53 PM, John Larkin wrote:
On Wed, 31 Jul 2019 11:54:42 -0400, bitrex <user@example.net> wrote:

On 7/31/19 11:40 AM, John Larkin wrote:
On Wed, 31 Jul 2019 11:21:47 -0400, bitrex <user@example.net> wrote:

On 7/31/19 10:22 AM, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier
operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load
resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.





Some probably do, but I've noticed a lot of discrete transistor data
sheets suck just as hard as any other kind of data sheet. I've never
seen a Smith chart on one...

e.g. like this is all ya get for a transistor array that costs ten
bucks?

https://www.mouser.com/datasheet/2/698/hfa3046-3096-3127-3128-1528732.pdf


It has a big table of s-params.

unfortunately the S parameter table and the charts are at two different
VCEs, idk why you would use this particular transistor in a 5 volt
system it seems clearly designed for low voltage/low power applications

also you don't get anything on noise figure vs collector current

here is a good data sheet:

http://www.hep.upenn.edu/SNO/daq/parts/mmbr941lt1.pdf

Is that part still available?

Most fast bipolars are gone now.




Here's a current manufacture one with all the nice data, and looks like
a Gummel-Poon spice model is available, too:

https://www.nxp.com/docs/en/data-sheet/BFU520.pdf

Looks like they are about 8 cent in quantities of 1k

 

[John Larkin]

On Wed, 31 Jul 2019 17:41:39 -0400, bitrex <user@example.net> wrote:

On 7/31/19 5:40 PM, bitrex wrote:
On 7/31/19 1:53 PM, John Larkin wrote:
On Wed, 31 Jul 2019 11:54:42 -0400, bitrex <user@example.net> wrote:

On 7/31/19 11:40 AM, John Larkin wrote:
On Wed, 31 Jul 2019 11:21:47 -0400, bitrex <user@example.net> wrote:

On 7/31/19 10:22 AM, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier
operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load
resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.





Some probably do, but I've noticed a lot of discrete transistor data
sheets suck just as hard as any other kind of data sheet. I've never
seen a Smith chart on one...

e.g. like this is all ya get for a transistor array that costs ten
bucks?

https://www.mouser.com/datasheet/2/698/hfa3046-3096-3127-3128-1528732.pdf


It has a big table of s-params.

unfortunately the S parameter table and the charts are at two different
VCEs, idk why you would use this particular transistor in a 5 volt
system it seems clearly designed for low voltage/low power applications

also you don't get anything on noise figure vs collector current

here is a good data sheet:

http://www.hep.upenn.edu/SNO/daq/parts/mmbr941lt1.pdf

Is that part still available?

Most fast bipolars are gone now.




Here's a current manufacture one with all the nice data, and looks like
a Gummel-Poon spice model is available, too:

https://www.nxp.com/docs/en/data-sheet/BFU520.pdf

Looks like they are about 8 cent in quantities of 1k

That's nice, 24 volts and 10 GHz. The pinout is bizarre.


The RF PNP situation remains grim.




--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[John Larkin]

On Wed, 31 Jul 2019 17:25:25 -0400, bitrex <user@example.net> wrote:

On 7/31/19 1:47 PM, John Larkin wrote:
On Wed, 31 Jul 2019 13:13:12 -0400, bitrex <user@example.net> wrote:

On 7/31/19 12:59 PM, tabbypurr@gmail.com wrote:
On Wednesday, 31 July 2019 16:01:42 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 07:52:05 -0700 (PDT), tabbypurr wrote:
On Wednesday, 31 July 2019 15:23:01 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.

Running Spice might be easier for the OP


NT

The problem there is that the RF people rarely provide Spice models,
and I suspect that the Spice models for most transistors are
inaccurate at high frequencies. For example, they typically don't
include wire bond inductance.

I find them inaccurate at low frequencies. When I use Spice it's as often as not to model taking into account all those edge situations eg varying Vbe, beta, supply line variation & so on, and Spice is no star at that level of detail. But it helps me clarify what's going on and I can work out the realities then.

But as a general suggestion, the OP will learn a ton by getting Spice running.


NT


if u think the BJT models are bad try the JFET models sometime...

Some jfets have a 10:1 range of Idss on the data sheet.



If the choice is between using a discrete JFET in a design and eating
soap It's usually a toss-up except, if the JFET is used as a diode.

Jfets are terrible diodes. Lots of series resistance, lots of
capacitance, a relative lot of reverse leakage. The c-b junction of a
cheap RF transistor is a vastly better diode.


--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[bitrex]

On 7/31/19 7:55 PM, John Larkin wrote:

On Wed, 31 Jul 2019 17:25:25 -0400, bitrex <user@example.net> wrote:

On 7/31/19 1:47 PM, John Larkin wrote:
On Wed, 31 Jul 2019 13:13:12 -0400, bitrex <user@example.net> wrote:

On 7/31/19 12:59 PM, tabbypurr@gmail.com wrote:
On Wednesday, 31 July 2019 16:01:42 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 07:52:05 -0700 (PDT), tabbypurr wrote:
On Wednesday, 31 July 2019 15:23:01 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.

Running Spice might be easier for the OP


NT

The problem there is that the RF people rarely provide Spice models,
and I suspect that the Spice models for most transistors are
inaccurate at high frequencies. For example, they typically don't
include wire bond inductance.

I find them inaccurate at low frequencies. When I use Spice it's as often as not to model taking into account all those edge situations eg varying Vbe, beta, supply line variation & so on, and Spice is no star at that level of detail. But it helps me clarify what's going on and I can work out the realities then.

But as a general suggestion, the OP will learn a ton by getting Spice running.


NT


if u think the BJT models are bad try the JFET models sometime...

Some jfets have a 10:1 range of Idss on the data sheet.



If the choice is between using a discrete JFET in a design and eating
soap It's usually a toss-up except, if the JFET is used as a diode.

Jfets are terrible diodes. Lots of series resistance, lots of
capacitance, a relative lot of reverse leakage. The c-b junction of a
cheap RF transistor is a vastly better diode.

"That's the joke"

They make OK zener didoes in a pinch

 

[John Larkin]

On Wed, 31 Jul 2019 20:33:13 -0400, bitrex <user@example.net> wrote:

On 7/31/19 7:55 PM, John Larkin wrote:
On Wed, 31 Jul 2019 17:25:25 -0400, bitrex <user@example.net> wrote:

On 7/31/19 1:47 PM, John Larkin wrote:
On Wed, 31 Jul 2019 13:13:12 -0400, bitrex <user@example.net> wrote:

On 7/31/19 12:59 PM, tabbypurr@gmail.com wrote:
On Wednesday, 31 July 2019 16:01:42 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 07:52:05 -0700 (PDT), tabbypurr wrote:
On Wednesday, 31 July 2019 15:23:01 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.

Running Spice might be easier for the OP


NT

The problem there is that the RF people rarely provide Spice models,
and I suspect that the Spice models for most transistors are
inaccurate at high frequencies. For example, they typically don't
include wire bond inductance.

I find them inaccurate at low frequencies. When I use Spice it's as often as not to model taking into account all those edge situations eg varying Vbe, beta, supply line variation & so on, and Spice is no star at that level of detail. But it helps me clarify what's going on and I can work out the realities then.

But as a general suggestion, the OP will learn a ton by getting Spice running.


NT


if u think the BJT models are bad try the JFET models sometime...

Some jfets have a 10:1 range of Idss on the data sheet.



If the choice is between using a discrete JFET in a design and eating
soap It's usually a toss-up except, if the JFET is used as a diode.

Jfets are terrible diodes. Lots of series resistance, lots of
capacitance, a relative lot of reverse leakage. The c-b junction of a
cheap RF transistor is a vastly better diode.



"That's the joke"

They make OK zener didoes in a pinch

Some people don't get the joke:

https://www.mouser.com/ProductDetail/InterFET/PAD1?qs=OxRSArmBDfzNntVAJAN1dw%3D%3D




--

John Larkin Highland Technology, Inc
picosecond timing precision measurement

jlarkin att highlandtechnology dott com
http://www.highlandtechnology.com

 

[bitrex]

On 7/31/19 8:52 PM, John Larkin wrote:

If the choice is between using a discrete JFET in a design and eating
soap It's usually a toss-up except, if the JFET is used as a diode.

Jfets are terrible diodes. Lots of series resistance, lots of
capacitance, a relative lot of reverse leakage. The c-b junction of a
cheap RF transistor is a vastly better diode.



"That's the joke"

They make OK zener didoes in a pinch

Some people don't get the joke:

https://www.mouser.com/ProductDetail/InterFET/PAD1?qs=OxRSArmBDfzNntVAJAN1dw%3D%3D

That's ah spicy-ah meat-ah ball!

$13 is a lot to pay for a forward voltage spec of: ah, it's, eh. whatever

 

[Phil Hobbs]

On 7/31/19 1:47 PM, John Larkin wrote:

On Wed, 31 Jul 2019 13:13:12 -0400, bitrex <user@example.net> wrote:

On 7/31/19 12:59 PM, tabbypurr@gmail.com wrote:
On Wednesday, 31 July 2019 16:01:42 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 07:52:05 -0700 (PDT), tabbypurr wrote:
On Wednesday, 31 July 2019 15:23:01 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.

Running Spice might be easier for the OP


NT

The problem there is that the RF people rarely provide Spice models,
and I suspect that the Spice models for most transistors are
inaccurate at high frequencies. For example, they typically don't
include wire bond inductance.

I find them inaccurate at low frequencies. When I use Spice it's as often as not to model taking into account all those edge situations eg varying Vbe, beta, supply line variation & so on, and Spice is no star at that level of detail. But it helps me clarify what's going on and I can work out the realities then.

But as a general suggestion, the OP will learn a ton by getting Spice running.


NT


if u think the BJT models are bad try the JFET models sometime...

Some jfets have a 10:1 range of Idss on the data sheet.

I'm doing a front end for a tiled MPPC (silicon PMT) that needs 40 MHz
bandwidth and has a capacitance of 700 pF. I'm going to use a SAV-551+
bootstrap, but Mini Circuits doesn't have a SPICE model for it, so I
used the one for the ATF38143 of fond memory.

I can calculate the noise easily, but just for grins I did a SPICE
..noise run. The pHEMT, running at 10 mA from a 4V supply, allegedly
will produce 74.2 kV RMS noise. I'm rich!

<https://electrooptical.net/www/sed/KilovoltNoise.png>

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[Cursitor Doom]

On Wed, 31 Jul 2019 10:47:05 -0700, John Larkin wrote:

> Some jfets have a 10:1 range of Idss on the data sheet.

Is LTSpice able to run monte carlos yet? I haven't had cause to use it
for about 2 years now.




--
This message may be freely reproduced without limit or charge only via
the Usenet protocol. Reproduction in whole or part through other
protocols, whether for profit or not, is conditional upon a charge of
GBP10.00 per reproduction. Publication in this manner via non-Usenet
protocols constitutes acceptance of this condition.

 

On Thursday, 1 August 2019 14:59:15 UTC+1, Phil Hobbs wrote:

I'm doing a front end for a tiled MPPC (silicon PMT) that needs 40 MHz
bandwidth and has a capacitance of 700 pF. I'm going to use a SAV-551+
bootstrap, but Mini Circuits doesn't have a SPICE model for it, so I
used the one for the ATF38143 of fond memory.

I can calculate the noise easily, but just for grins I did a SPICE
.noise run. The pHEMT, running at 10 mA from a 4V supply, allegedly
will produce 74.2 kV RMS noise. I'm rich!

https://electrooptical.net/www/sed/KilovoltNoise.png

Cheers

Phil Hobbs

74kV noise on a 12v supply? Sounds about right lol.


NT

 

[John Larkin]

On Thu, 1 Aug 2019 09:59:02 -0400, Phil Hobbs
<pcdhSpamMeSenseless@electrooptical.net> wrote:

On 7/31/19 1:47 PM, John Larkin wrote:
On Wed, 31 Jul 2019 13:13:12 -0400, bitrex <user@example.net> wrote:

On 7/31/19 12:59 PM, tabbypurr@gmail.com wrote:
On Wednesday, 31 July 2019 16:01:42 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 07:52:05 -0700 (PDT), tabbypurr wrote:
On Wednesday, 31 July 2019 15:23:01 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.

Running Spice might be easier for the OP


NT

The problem there is that the RF people rarely provide Spice models,
and I suspect that the Spice models for most transistors are
inaccurate at high frequencies. For example, they typically don't
include wire bond inductance.

I find them inaccurate at low frequencies. When I use Spice it's as often as not to model taking into account all those edge situations eg varying Vbe, beta, supply line variation & so on, and Spice is no star at that level of detail. But it helps me clarify what's going on and I can work out the realities then.

But as a general suggestion, the OP will learn a ton by getting Spice running.


NT


if u think the BJT models are bad try the JFET models sometime...

Some jfets have a 10:1 range of Idss on the data sheet.



I'm doing a front end for a tiled MPPC (silicon PMT) that needs 40 MHz
bandwidth and has a capacitance of 700 pF. I'm going to use a SAV-551+
bootstrap, but Mini Circuits doesn't have a SPICE model for it, so I
used the one for the ATF38143 of fond memory.

I can calculate the noise easily, but just for grins I did a SPICE
.noise run. The pHEMT, running at 10 mA from a 4V supply, allegedly
will produce 74.2 kV RMS noise. I'm rich!

https://electrooptical.net/www/sed/KilovoltNoise.png

Cheers

Phil Hobbs

Free energy. Make a Youtube video.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[John S]

On 8/1/2019 12:21 PM, Cursitor Doom wrote:

On Wed, 31 Jul 2019 10:47:05 -0700, John Larkin wrote:

Some jfets have a 10:1 range of Idss on the data sheet.

Is LTSpice able to run monte carlos yet? I haven't had cause to use it
for about 2 years now.

Look at "MonteCarlo.asc" in the Educational folder.

 

[John Larkin]

On Fri, 2 Aug 2019 06:59:22 -0500, John S <Sophi.2@invalid.org> wrote:

On 8/1/2019 12:21 PM, Cursitor Doom wrote:
On Wed, 31 Jul 2019 10:47:05 -0700, John Larkin wrote:

Some jfets have a 10:1 range of Idss on the data sheet.

Is LTSpice able to run monte carlos yet? I haven't had cause to use it
for about 2 years now.

Look at "MonteCarlo.asc" in the Educational folder.

My guys run MC sims in LT Spice. I don't think it can frob the
parameters internal to a device model, like Idss. Can it?


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

 

[bitrex]

On 7/31/19 7:59 PM, John Larkin wrote:

Here's a current manufacture one with all the nice data, and looks like
a Gummel-Poon spice model is available, too:

https://www.nxp.com/docs/en/data-sheet/BFU520.pdf

Looks like they are about 8 cent in quantities of 1k

That's nice, 24 volts and 10 GHz. The pinout is bizarre.


The RF PNP situation remains grim.

Looks like these are not discontinued but they'll make them for you if
you order a bunch:

<https://www.mouser.com/datasheet/2/68/cm4957-1149866.pdf>

 

[Phil Hobbs]

On 8/2/19 11:38 AM, John Larkin wrote:

On Fri, 2 Aug 2019 06:59:22 -0500, John S <Sophi.2@invalid.org> wrote:

On 8/1/2019 12:21 PM, Cursitor Doom wrote:
On Wed, 31 Jul 2019 10:47:05 -0700, John Larkin wrote:

Some jfets have a 10:1 range of Idss on the data sheet.

Is LTSpice able to run monte carlos yet? I haven't had cause to use it
for about 2 years now.

Look at "MonteCarlo.asc" in the Educational folder.

My guys run MC sims in LT Spice. I don't think it can frob the
parameters internal to a device model, like Idss. Can it?

Yes. You go
..model myfet ako:BF862 Idss=5mA

or
..model myfet ako:BF862 Idss={Id_at_zero_bias}
..step param Id_at_zero_bias list 8m 12m 17m

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

 

[amal banerjee]

On Wednesday, July 31, 2019 at 11:01:42 AM UTC-4, John Larkin wrote:

On Wed, 31 Jul 2019 07:52:05 -0700 (PDT), tabbypurr@gmail.com wrote:

On Wednesday, 31 July 2019 15:23:01 UTC+1, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.

Running Spice might be easier for the OP


NT

The problem there is that the RF people rarely provide Spice models,
and I suspect that the Spice models for most transistors are
inaccurate at high frequencies. For example, they typically don't
include wire bond inductance.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics

For new breed of RF transistors as HFA3134, small signal S parameters are not provided, and only the
detailed SPICE model is available.

 

[amal banerjee]

On Wednesday, July 31, 2019 at 10:23:01 AM UTC-4, John Larkin wrote:

On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.




--

John Larkin Highland Technology, Inc

lunatic fringe electronics

Yes, I have tried the simple formula
S11 = (Zin - Z0)/(Zin + Z0), assuming a standard RF Z0=50.0 Ohm, for listed S parameter freqeuncies. for
non-listed frequencies, the small signal S parameters can be obtained via SPICE .AC analysis.

 

[amal banerjee]

On Wednesday, July 31, 2019 at 11:21:53 AM UTC-4, bitrex wrote:

On 7/31/19 10:22 AM, John Larkin wrote:
On Wed, 31 Jul 2019 03:25:52 -0700 (PDT), amal banerjee
dakupoto@gmail.com> wrote:

Could some electronics guru here please clarify this a bit ?
What is the input impedance for a common emitter BJT amplifier operating at RF frequencies ? The output impedance id the
parallel combination of the collector resistor Rc and the load resistor. All hints, suggestionas are welcome. Thanks in advance.


The data sheets of RF transistors usually have tables of s-parameters
or Smith charts.





Some probably do, but I've noticed a lot of discrete transistor data
sheets suck just as hard as any other kind of data sheet. I've never
seen a Smith chart on one...

e.g. like this is all ya get for a transistor array that costs ten bucks?

https://www.mouser.com/datasheet/2/698/hfa3046-3096-3127-3128-1528732.pdf

this is all ya get for a uhf power transistor that costs 3 bucks:

https://www.mouser.com/datasheet/2/302/BLT81_2-76125.pdf

So true. I have spent hours and hours digging out SPICE models for RF transistors, online.