|
Hi gang,? Not complaining ...? ?Just curious.
I had recently built 3 of the 50W amps ( 15M, 20M and 40M). Want to use them for a little extra punch (say 10 to no more than 20 watts out) for field semi-portable 'digi' work (FT8, J8Call, WSPR) . . .? well, just because.??I built all three as '12 volt' versions.?
?I set the drive bias exactly as described in the instructions. The builds went well and the amps appear to output as described in the manual(s)./graphs.
I was just wondering why the 40 meter version is 'hotter' than say the 20 meter version which is hotter than the 15 meter version.
I found that I can get 10 Watts OUT with the following Watts IN from my K2.
15M - approx 4.5 Watts IN
20M - approx 2 Watts IN
40M -? LOTS of WATTS OUT (Too Many) with Just 1 Watt in !!
** For the 40M, I found if I backed down the bias, I could get 10 Watts out by setting the K2 for between 0.1 and 0.2 Watts out. Yep 'tenths' of a Watt.?
Is my stumbling on the bias backdown a potential problem point?
Or, perhaps I should make up some PI attenuator circuits to add to the Transmit circuit chains in the AMPs where needed to tone down the 3.6 - 3.9 Watts that my new QMX is producing??
Thanks in advance,?
Greg? KI4NVX
|
Hi- is this into a watt/swr meter then dummy load?? Antenna connections with SWR issues can fool inexpensive watt meters.?
Larry KB3CUF?
On Wed, Mar 13, 2024 at 11:23?PM Greg McCain via <gamccain50= [email protected]> wrote: Hi gang,? Not complaining ...? ?Just curious.
I had recently built 3 of the 50W amps ( 15M, 20M and 40M). Want to use them for a little extra punch (say 10 to no more than 20 watts out) for field semi-portable 'digi' work (FT8, J8Call, WSPR) . . .? well, just because.??I built all three as '12 volt' versions.?
?I set the drive bias exactly as described in the instructions. The builds went well and the amps appear to output as described in the manual(s)./graphs.
I was just wondering why the 40 meter version is 'hotter' than say the 20 meter version which is hotter than the 15 meter version.
I found that I can get 10 Watts OUT with the following Watts IN from my K2.
15M - approx 4.5 Watts IN
20M - approx 2 Watts IN
40M -? LOTS of WATTS OUT (Too Many) with Just 1 Watt in !!
** For the 40M, I found if I backed down the bias, I could get 10 Watts out by setting the K2 for between 0.1 and 0.2 Watts out. Yep 'tenths' of a Watt.?
Is my stumbling on the bias backdown a potential problem point?
Or, perhaps I should make up some PI attenuator circuits to add to the Transmit circuit chains in the AMPs where needed to tone down the 3.6 - 3.9 Watts that my new QMX is producing??
Thanks in advance,?
Greg? KI4NVX
|
Yes.
K2 - 50W Amp - QrpGuys SWR/Power meter.
On Thu, Mar 14, 2024 at 12:29 AM, Larry Acklin <acklin@...> wrote: Hi- is this into a watt/swr meter then dummy load?? Antenna connections with SWR issues can fool inexpensive watt meters.?
Larry KB3CUF? On Wed, Mar 13, 2024 at 11:23?PM Greg McCain via <gamccain50= [email protected]> wrote: Hi gang,? Not complaining ...? ?Just curious.
I had recently built 3 of the 50W amps ( 15M, 20M and 40M). Want to use them for a little extra punch (say 10 to no more than 20 watts out) for field semi-portable 'digi' work (FT8, J8Call, WSPR) . . .? well, just because.??I built all three as '12 volt' versions.?
?I set the drive bias exactly as described in the instructions. The builds went well and the amps appear to output as described in the manual(s)./graphs.
I was just wondering why the 40 meter version is 'hotter' than say the 20 meter version which is hotter than the 15 meter version.
I found that I can get 10 Watts OUT with the following Watts IN from my K2.
15M - approx 4.5 Watts IN
20M - approx 2 Watts IN
40M -? LOTS of WATTS OUT (Too Many) with Just 1 Watt in !!
** For the 40M, I found if I backed down the bias, I could get 10 Watts out by setting the K2 for between 0.1 and 0.2 Watts out. Yep 'tenths' of a Watt.?
Is my stumbling on the bias backdown a potential problem point?
Or, perhaps I should make up some PI attenuator circuits to add to the Transmit circuit chains in the AMPs where needed to tone down the 3.6 - 3.9 Watts that my new QMX is producing??
Thanks in advance,?
Greg? KI4NVX
|
When you see an amplifier provide loads of gain in a particular band or frequency it is usually because there is a circuit self resonance, so losses are minimal, or there is some negative feedback that due to phase shifts is becoming less so. ?It is not much fun when the amplifier gain and phase shifts flip into the positive region and oscillation begins.
toggle quoted message
Show quoted text
On Mar 14, 2024, at 07:13, Greg McCain via groups.io <gamccain50@...> wrote:
?Yes.
K2 - 50W Amp - QrpGuys SWR/Power meter.
Greg On Thu, Mar 14, 2024 at 12:29 AM, Larry Acklin <acklin@...> wrote: Hi- is this into a watt/swr meter then dummy load?? Antenna connections with SWR issues can fool inexpensive watt meters.?
Larry KB3CUF? On Wed, Mar 13, 2024 at 11:23?PM Greg McCain via <gamccain50= [email protected]> wrote: Hi gang,? Not complaining ...? ?Just curious.
I had recently built 3 of the 50W amps ( 15M, 20M and 40M). Want to use them for a little extra punch (say 10 to no more than 20 watts out) for field semi-portable 'digi' work (FT8, J8Call, WSPR) . . .? well, just because.??I built all three as '12 volt' versions.?
?I set the drive bias exactly as described in the instructions. The builds went well and the amps appear to output as described in the manual(s)./graphs.
I was just wondering why the 40 meter version is 'hotter' than say the 20 meter version which is hotter than the 15 meter version.
I found that I can get 10 Watts OUT with the following Watts IN from my K2.
15M - approx 4.5 Watts IN
20M - approx 2 Watts IN
40M -? LOTS of WATTS OUT (Too Many) with Just 1 Watt in !!
** For the 40M, I found if I backed down the bias, I could get 10 Watts out by setting the K2 for between 0.1 and 0.2 Watts out. Yep 'tenths' of a Watt.?
Is my stumbling on the bias backdown a potential problem point?
Or, perhaps I should make up some PI attenuator circuits to add to the Transmit circuit chains in the AMPs where needed to tone down the 3.6 - 3.9 Watts that my new QMX is producing??
Thanks in advance,?
Greg? KI4NVX
|
It¡¯s possible the 40M amp is oscillating. Also possible that the input Pi attenuaor has an issue.?
I have 2 of those 40M amps and one gets 25W with 4 W in at 13V
2nd gets 50W with 4W in at 22V
Larry KB3CUF?
On Thu, Mar 14, 2024 at 7:13?AM Greg McCain via <gamccain50= [email protected]> wrote: Yes.
K2 - 50W Amp - QrpGuys SWR/Power meter.
Greg On Thu, Mar 14, 2024 at 12:29 AM, Larry Acklin Hi- is this into a watt/swr meter then dummy load?? Antenna connections with SWR issues can fool inexpensive watt meters.?
Larry KB3CUF? On Wed, Mar 13, 2024 at 11:23?PM Greg McCain via <gamccain50= [email protected]> wrote: Hi gang,? Not complaining ...? ?Just curious.
I had recently built 3 of the 50W amps ( 15M, 20M and 40M). Want to use them for a little extra punch (say 10 to no more than 20 watts out) for field semi-portable 'digi' work (FT8, J8Call, WSPR) . . .? well, just because.??I built all three as '12 volt' versions.?
?I set the drive bias exactly as described in the instructions. The builds went well and the amps appear to output as described in the manual(s)./graphs.
I was just wondering why the 40 meter version is 'hotter' than say the 20 meter version which is hotter than the 15 meter version.
I found that I can get 10 Watts OUT with the following Watts IN from my K2.
15M - approx 4.5 Watts IN
20M - approx 2 Watts IN
40M -? LOTS of WATTS OUT (Too Many) with Just 1 Watt in !!
** For the 40M, I found if I backed down the bias, I could get 10 Watts out by setting the K2 for between 0.1 and 0.2 Watts out. Yep 'tenths' of a Watt.?
Is my stumbling on the bias backdown a potential problem point?
Or, perhaps I should make up some PI attenuator circuits to add to the Transmit circuit chains in the AMPs where needed to tone down the 3.6 - 3.9 Watts that my new QMX is producing??
Thanks in advance,?
Greg? KI4NVX
|
Hi Greg,
AFAIK, IRF510 was designed to be a power switching device. Many years ago, some kind soul found that it can also be used as a low cost HF power amp, with more than decent results. But, as there's no free lunch in this universe, so is the case with IRF510 - due to it's internal capacities, it works best in the low bands and kinda so-so under 20m.
Incidentally, my first amp that didn't oscillate and actually worked was a IRF510 design for 80 and 40 meter, published by a DL ham. It was 25 years ago, I think I still have it somewhere under the bed :-)?
Using some sort of resistive attenuator is probably a good idea, if you can't power the QMX from something lower than 12V. While tinkering with atten. and bias, it would be nice to have a scope to make sure that you actually output a sine.
Regards,
YO3GFH
op. Adrian
|
I have a linear amp that gets over 40W at 6M using IRF510s.
Wasn't hard to do that.
The characteristics are not great but its not about capacitance as
a sole factor.? Then again input Z is a factor even MRF and BLF series
MOSFETs.
The problem of wideband amps with MOSFETs is input and output matching.
The gate is a capacitor and that makes input matching difficult.? Most designs
using IRF510 or IRF530 40M is a sweet spot.? The problem being input
impedance goes down with increasing frequency.
Develop voltage at the gate effectively you get power out.
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.
|
Yes, 100% agreed with Allison.?
MOSFETs don't have an intrinsic frequency limit like a bipolar. Which is also why good board layout and attention to parasitics is important because you can easily get VHF oscillations.?
But as Allison said the gate capacitance is the issue. Which in other words means it gets harder and harder to match the ever lower impedance efficiently as you go up in frequency.?
Even though the IRF510 is often accused of only being a car headlamp switching transistor... These capacitance and impedance issues affect even "real" RF MOSFETs as much or almost as much.?
With careful feedback and layout as in the 10W Linear amp kit the gain can be made quite flat across HF 1.8-30MHz to within a dB or two.?
The 50W PA kit was a much more basic design effort originally done for 40m where everything is easiest as Allison mentioned. Reasonable worthwhile performance obtained up to 20m made those acceptable options too. I'm not surprised 15m was even lower down.?
73 Hans G0UPL
On Thu, Mar 14, 2024, 7:30?PM ajparent1/kb1gmx < kb1gmx@...> wrote: I have a linear amp that gets over 40W at 6M using IRF510s.
Wasn't hard to do that.
The characteristics are not great but its not about capacitance as
a sole factor.? Then again input Z is a factor even MRF and BLF series
MOSFETs.
The problem of wideband amps with MOSFETs is input and output matching.
The gate is a capacitor and that makes input matching difficult.? Most designs
using IRF510 or IRF530 40M is a sweet spot.? The problem being input
impedance goes down with increasing frequency.
Develop voltage at the gate effectively you get power out.
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.
|
15M? results say the amp has issues. Either low bias or LPF filter??
20M 10W for 2 in seems? a bit low as well (I'D expect 13-15W)
At the same time the 40M results suggests maybe other things going on.
For 40M at 3W in I'd expect about 17-25W at 13.8V.? Part of the reason
is there was no intention to get flat gain from the amp.? No surprise with
a class C amp.? Hoeever if the bias is set high you do get more power
and possibly other things like spurs.
The IRF510 is performs poorly at lower voltages that why at 20V it can easily do 50W
but at 13.8 you may get around 20W.
Oddly I have an amp that uses MRF141(dual) and at 28V its easily 330W, but at 13.8V
it barely does 90W for the same drive.? THis is often seen with all Mosfets and Bipolar
devices because whne you lower votlages characteristics change.
--
Allison
------------------
Post online only,?
direct email will go to a bit bucket.
|
Thx Adrian, Hans and Allison.
Appreciate the feedback.
Will proceed carefully.
Greg
On Thu, Mar 14, 2024 at 9:49 AM, Adrian YO3GFH via groups.io <p_adr@...> wrote: Hi Greg,
AFAIK, IRF510 was designed to be a power switching device. Many years ago, some kind soul found that it can also be used as a low cost HF power amp, with more than decent results. But, as there's no free lunch in this universe, so is the case with IRF510 - due to it's internal capacities, it works best in the low bands and kinda so-so under 20m.
Incidentally, my first amp that didn't oscillate and actually worked was a IRF510 design for 80 and 40 meter, published by a DL ham. It was 25 years ago, I think I still have it somewhere under the bed :-)?
Using some sort of resistive attenuator is probably a good idea, if you can't power the QMX from something lower than 12V. While tinkering with atten. and bias, it would be nice to have a scope to make sure that you actually output a sine.
Regards,
YO3GFH
op. Adrian
|
For an output power of 40 Watt, the input power was: 80 m - 0.4 Watt; 40 m - 1.2 Watt; 30 m - 1.5 Watt; 20 m - 2.2 Watt.
The amplifier supply voltage is 24 V, the current consumption is about 3 A. The input power was changed by changing the supply voltage of the output stage QDX
-- 73, Oleh, UY2ZA
|
