|
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ?? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
Sure, a couple of possibilities:
1) When your power supply is off it's a high impedance, and when it's on it's a low impedance, so you are at the very least switching in another inductor.
2) Semiconductor junction capacitance depends on applied voltage, and there are semiconductors in the amplifier block.
3) Your amplifier block is accepting input power (to amplify it) or not. You could think of this as an additonal load.
Without more details we can only speculate, but I'm not surprised at your results. I would certainly expect the input of the amplifier to change when it's powered up.
73, Willie N1JBJ
toggle quoted message
Show quoted text
On Jun 28, 2021, at 4:02 AM, Peter <yliroma@...> wrote:
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ?? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
<Match_Meas.JPG>
|
Hi Peter,
every amplifier (Transistor or Tube) will change its input (and output) impedance due to bias. So input (and output) impedance of impeadance matching network as shown in your picture will also change when bias to the amplifier is supplied.
73, Guenter, DK5DN
Am 28.06.2021 um 10:02 schrieb Peter:
toggle quoted message
Show quoted text
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ?? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
Peter, you haven't indicated whether the amplifier is linear (class A or AB or B) or class C (for FM or CW). The circuit operating conditions (class, signal level, bias, Vcc) make a big difference. Input match The input match is generally designed to present 50¦¸ to the input socket. Output match The output match is generally designed to deliver maximum power to a 50¦¸ load. The PA is *not* a conjugate match, which would reduce the output power by 6dB! Take care A nanoVNA (most VNAs for that matter) is not safe anywhere near a 200W amplifier. All it would take is a spurious oscillation in the PA, and the VNA is likely to see rather more than +10dBm (10mW) at a measuring port, with the danger of damaging the VNA electronics. (See also /g/nanovna-users/topic/77980973#18555 < /g/nanovna-users/topic/77980973#18555>, for instance.) HTH, 73, Stay Safe, Robin, G8DQX
On 28/06/2021 12:38, schweppe wrote: Hi Peter,
every amplifier (Transistor or Tube) will change its input (and output) impedance due to bias. So input (and output) impedance of impeadance matching network as shown in your picture will also change when bias to the amplifier is supplied.
73, Guenter, DK5DN
Am 28.06.2021 um 10:02 schrieb Peter:
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value? associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ???? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
On 6/28/21 5:43 AM, G8DQX list wrote: Peter,
you haven't indicated whether the amplifier is linear (class A or AB or B) or class C (for FM or CW). The circuit operating conditions (class, signal level, bias, Vcc) make a big difference.
Input match
The input match is generally designed to present 50¦¸ to the input socket.
Output match
The output match is generally designed to deliver maximum power to a 50¦¸ load. The PA is *not* a conjugate match, which would reduce the output power by 6dB!
Take care
A nanoVNA (most VNAs for that matter) is not safe anywhere near a 200W amplifier. All it would take is a spurious oscillation in the PA, and the VNA is likely to see rather more than +10dBm (10mW) at a measuring port, with the danger of damaging the VNA electronics. (See also /g/nanovna-users/topic/77980973#18555 </g/nanovna-users/topic/77980973#18555>, for instance.) If it breaks into oscillation (or picks up a spurious signal) it could easily push more than 10dBm *out* of the input port. Even operating as designed, it could have surprisingly high power coming out the input. In order to be stable, the S12 (reverse isolation) just has to be bigger than the forward gain (S21) (slight oversimplification). Let's say the amplifier gain is 10 dB,? S12 is -15 dB, and a mismatched load reflects -20dB, and the amplifier is putting out 100W (+50dBm) - the reflected power from the load is +30dBm, the amplifier (going backwards) reduces that to +15dBm.
|
Thanks to eveyone gave an answer ...
Of course I realize the Bias is affecting the Input (and output) impedance of any device ... BUT WHY ???
I intend to highlight my P.A. is a A/B Class RF Amplifier equipeb by two parallele transistor coupled somehow:
Attached the schematics : it is a surplus (ex milititary) unit which was working originally in AM mode ... the related
Bias is obtained via a paritcular DIODE ...
For those who think DC Power-Supply introduces some extra impedance: this not true .. all +Vcc Supply lines are grounded by the CPs capacitors and RF is blocked by the JAFs...
See the attached schematics.
At any rate I have to connect my NANO-VNA when the system is Power-supplied ... ISN'T IT ???
Peter
toggle quoted message
Show quoted text
----- Original Message ----- From: "schweppe" <schweppe@...> To: < [email protected]> Sent: Monday, June 28, 2021 1:38 PM Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications Hi Peter,
every amplifier (Transistor or Tube) will change its input (and output)
impedance due to bias. So input (and output) impedance of impeadance
matching network as shown in your picture will also change when bias to
the amplifier is supplied.
73, Guenter, DK5DN
Am 28.06.2021 um 10:02 schrieb Peter:
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ?? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
Aside: not sure why you are biasing the input, looks like the transistors are already biased on thru BYI-1F, and the input is capacitively coupled.
At the very least the B-E junction of a transistor changes its¡¯ capacitance when you bias it on, no?
William Smith
ComputerSmiths Consulting, Inc.
w_smith@...
toggle quoted message
Show quoted text
On Jun 28, 2021, at 9:57 AM, Peter <yliroma@...> wrote:
?Thanks to eveyone gave an answer ...
Of course I realize the Bias is affecting the Input (and output) impedance of any device ... BUT WHY ???
I intend to highlight my P.A. is a A/B Class RF Amplifier equipeb by two parallele transistor coupled somehow:
Attached the schematics : it is a surplus (ex milititary) unit which was working originally in AM mode ... the related
Bias is obtained via a paritcular DIODE ...
For those who think DC Power-Supply introduces some extra impedance: this not true .. all +Vcc Supply lines are grounded by the CPs capacitors and RF is blocked by the JAFs...
See the attached schematics.
At any rate I have to connect my NANO-VNA when the system is Power-supplied ... ISN'T IT ???
Peter
----- Original Message ----- From: "schweppe" <schweppe@...>
To: <[email protected]>
Sent: Monday, June 28, 2021 1:38 PM
Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications
Hi Peter,
every amplifier (Transistor or Tube) will change its input (and output)
impedance due to bias. So input (and output) impedance of impeadance
matching network as shown in your picture will also change when bias to
the amplifier is supplied.
73, Guenter, DK5DN
Am 28.06.2021 um 10:02 schrieb Peter:
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ?? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
Hi William ... of course the "squared BIAS" I Drawn is just the one made with the BY1-1F ... there is no any other Bias ...
I was asking WHY the INPUT Resonance changed as soon as I apply Power-Supply (and consequentely the Bias) ....
I was just a question ...
Peter.
toggle quoted message
Show quoted text
----- Original Message ----- From: "William Smith" <w_smith@...> To: < [email protected]> Sent: Monday, June 28, 2021 5:46 PM Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications Aside: not sure why you are biasing the input, looks like the transistors are already biased on thru BYI-1F, and the input is capacitively coupled. At the very least the B-E junction of a transistor changes its¡¯ capacitance when you bias it on, no? William Smith ComputerSmiths Consulting, Inc. w_smith@... On Jun 28, 2021, at 9:57 AM, Peter <yliroma@...> wrote:
?Thanks to eveyone gave an answer ...
Of course I realize the Bias is affecting the Input (and output) impedance of any device ... BUT WHY ???
I intend to highlight my P.A. is a A/B Class RF Amplifier equipeb by two parallele transistor coupled somehow:
Attached the schematics : it is a surplus (ex milititary) unit which was working originally in AM mode ... the related
Bias is obtained via a paritcular DIODE ...
For those who think DC Power-Supply introduces some extra impedance: this not true .. all +Vcc Supply lines are grounded by the CPs capacitors and RF is blocked by the JAFs...
See the attached schematics.
At any rate I have to connect my NANO-VNA when the system is Power-supplied ... ISN'T IT ???
Peter
----- Original Message ----- From: "schweppe" <schweppe@...>
To: <[email protected]>
Sent: Monday, June 28, 2021 1:38 PM
Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications
Hi Peter,
every amplifier (Transistor or Tube) will change its input (and output)
impedance due to bias. So input (and output) impedance of impeadance
matching network as shown in your picture will also change when bias to
the amplifier is supplied.
73, Guenter, DK5DN
Am 28.06.2021 um 10:02 schrieb Peter:
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ?? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
On Mon, Jun 28, 2021 at 08:52 AM, Peter wrote: I was asking WHY the INPUT Resonance changed as soon as I apply Power-Supply
(and consequentely the Bias) ....
I was just a question ...
When you apply bias to the transistors Q1 and Q2 the input impedance looking into the base of the transistors changes. This affects the overall input impedance of the network (including any observed resonances) seen at the junction of the 1 nF and 120 pF capacitors. The semi-rigid coax is attached at this point so the impedance seen at the input to this coax will also change. The input resonant circuit formed by the 47 ohm resistor, 330 nH inductor, 220 pF cap, 47 pF cap in parallel with this coax input impedance will now have a different resonant frequency. Roger
|
Peter,
it helps to think of the active devices, the PA power transistors, as horribly non-linear devices. The base-emitter junction is a forward biased diode, whose small-signal parameters vary over an RF cycle. The collector-base junction is a reverse biased diode (think varicap diode <>), whose capacitance (Ccb) depends on the instantaneous value of the collector-base voltage. That capacitance is then subject to the Miller effect ( <>), and can provide positive feedback from output to input.
Generally, the stage voltage gain is kept quite low to avoid any danger of the apparatus becoming a power oscillator rather than a power amplifier.
And since the base emitter junction is a diode, it obeys (more or less) the diode equation for AC resistance, r'= .025/Ib?¦¸, where Ib is the base current. There is a lot more going on, for instance see <>.
All that capacitance has an impact on the tuning of the input and output networks. At this point, you might be thinking that any functioning RF amplifier is a triumph of engineering ingenuity over the Laws of Physics. You would not be far wrong.
73, Stay Safe,
Robin, G8DQX
PS: It was a good question, but with a complex answer, depending on how much detail one can cope with!
toggle quoted message
Show quoted text
On 28/06/2021 16:52, Peter wrote: Hi William ... of course the "squared BIAS" I Drawn is just the one made with the BY1-1F ... there is no any other Bias ...
I was asking WHY the INPUT Resonance changed as soon as I apply Power-Supply (and consequentely the Bias) ....
I was just a question ...
Peter.
----- Original Message ----- From: "William Smith" <w_smith@...>
To: <[email protected]>
Sent: Monday, June 28, 2021 5:46 PM
Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications
Aside:? not sure why you are biasing the input, looks like the transistors are already biased on thru BYI-1F, and the input is capacitively coupled.
At the very least the B-E junction of a transistor changes its¡¯ capacitance when you bias it on, no?
William Smith
ComputerSmiths Consulting, Inc.
w_smith@...
On Jun 28, 2021, at 9:57 AM, Peter <yliroma@...> wrote:
?Thanks to eveyone gave an answer ...
Of course I realize the Bias is affecting the Input (and output) impedance of any device ... BUT WHY ???
I intend to highlight my P.A. is a A/B Class RF Amplifier equipeb by two parallele transistor coupled somehow:
Attached the schematics : it is a surplus (ex milititary) unit which was working originally in AM mode ... the related
Bias is obtained via a paritcular DIODE ...
For those who think DC Power-Supply introduces some extra impedance: this not true .. all +Vcc Supply lines are grounded by the CPs capacitors and RF is blocked by the JAFs...
See the attached schematics.
At any rate I have to connect my NANO-VNA when the system is Power-supplied ... ISN'T IT ???
Peter
----- Original Message ----- From: "schweppe" <schweppe@...>
To: <[email protected]>
Sent: Monday, June 28, 2021 1:38 PM
Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications
Hi Peter,
every amplifier (Transistor or Tube) will change its input (and output)
impedance due to bias. So input (and output) impedance of impeadance
matching network as shown in your picture will also change when bias to
the amplifier is supplied.
73, Guenter, DK5DN
Am 28.06.2021 um 10:02 schrieb Peter:
Hi there ...
Appreciating several features of this incredible instrument, I'm attempting to check the real Impedance of the Matching network INPUT (and/or out-put) value? associated to a VHF RF Amplifier (medium-power abt 200W) ... I connected the VNA as shown on the picture and then, after related calibration, I perfomed the measure (sweep)....
Of course I didn't apply any RF source at input ... Nano-Vna only !
I found a strange Resonant Frequency ... quite far from the proper one (I expected to be around 140MHz...150MHz) .... But then , if I apply the related Power-Supply (including BIAS) , the situation changed ... the resonant frequency was really close to the expected one !
Is it normal ???? Really the Power-Supply/Bias could affect impedance values of Input matching net-work ?? This situation can be applied at Out-Put line as well ??
|
Many thanks Roger, for your answer ...
Just pointing-out to the series resonant circuit (at very input before the semi-rigid line) : 47 ohm - 330nH - 220pF !
I cannot understand what is its purpose ... it looks to be resonant around 18.6 MHz ... ??? ...
Is it a filter ? Why it is there ? This one is not my project ... it comes from military labs ...
One thing I realize for sure : INPUT (and Output) impedance measurementmade with Nano-Vna MUST be performed
having Power-Supply & Bias switched-ON !
Thanks to everybody !
Peter
toggle quoted message
Show quoted text
----- Original Message ----- From: "Roger Need via groups.io" <sailtamarack@...> To: < [email protected]> Sent: Monday, June 28, 2021 6:36 PM Subject: Re: [nanovna-users] I/O RF Matching net-works measuring with NANO-VNA #applications On Mon, Jun 28, 2021 at 08:52 AM, Peter wrote: I was asking WHY the INPUT Resonance changed as soon as I apply Power-Supply
(and consequentely the Bias) ....
I was just a question ...
When you apply bias to the transistors Q1 and Q2 the input impedance looking into the base of the transistors changes. This affects the overall input impedance of the network (including any observed resonances) seen at the junction of the 1 nF and 120 pF capacitors. The semi-rigid coax is attached at this point so the impedance seen at the input to this coax will also change. The input resonant circuit formed by the 47 ohm resistor, 330 nH inductor, 220 pF cap, 47 pF cap in parallel with this coax input impedance will now have a different resonant frequency. Roger
|
Robin, truly excellent teaching!
Thanks for taking the time to illuminate.
John
at radio station VE7AOV
+++++ On 2021-06-28 1:46 p.m., G8DQX list wrote: Peter,
it helps to think of the active devices, the PA power transistors, as
horribly non-linear devices. The base-emitter junction is a forward
biased diode, whose small-signal parameters vary over an RF cycle. The
collector-base junction is a reverse biased diode (think varicap diode
<>), whose capacitance (Ccb)
depends on the instantaneous value of the collector-base voltage. That
capacitance is then subject to the Miller effect
(
<>), and can provide
positive feedback from output to input.
Generally, the stage voltage gain is kept quite low to avoid any
danger of the apparatus becoming a power oscillator rather than a
power amplifier.
And since the base emitter junction is a diode, it obeys (more or
less) the diode equation for AC resistance, r'= .025/Ib?¦¸, where Ib is
the base current. There is a lot more going on, for instance see
<>.
All that capacitance has an impact on the tuning of the input and
output networks. At this point, you might be thinking that any
functioning RF amplifier is a triumph of engineering ingenuity over
the Laws of Physics. You would not be far wrong.
73, Stay Safe,
Robin, G8DQX
PS: It was a good question, but with a complex answer, depending on
how much detail one can cope with!
...
--
|
Peter
schweppe
Roger Need
