Hi all

It has been mentioned on this group a number of times in this discussion group so I am posting details here.?

1. I have now made available the QRP Labs 10W HF Linear PA kit, providing 26dB of gain with +/-1dB gain flatness from 2-30MHz, 10W output all the way up to 10m band from ?supply. This amplifier uses two BS170 in push-pull as the driver and two IRF510 in push-pull as the final stage. This kit shows that two IRF510 MOSFETs can produce excellent results in a high performance yet low cost ($26) linear amplifier kit.?

See for details.?

I am not suggesting that this amplifier will solve all problems and without wishing to be offensive, the "garbage in, garbage out" rule always applies - in particular, if you feed it a signal containing spurs, harmonics, or other problems... then you will get the same thing at the other end, just amplified 26dB. It also still applies that all stages leading up to the final amplifier must have flat gain. If the pre-drivers cannot provide a big enough signal then you won't get full power output.?

But, it could be useful or interesting, in the BITX context, to look at the details of this amplifier kit and see the short lead lengths, symmetric PCB and extensively "stitched" groundplanes.?

I am very grateful to Allison KB1GMX for very valuable advice during the design and development of this amplifier, beta-testing prototypes, measurements and moral support.

73 Hans G0UPL

> Congrats for the amplifier. Surely interesting for TX homebrewing?

> at a reasonable cost. Is there any evidence that it would improve?

> the PA of the uBitx?

Evidence, no. I don't have a uBITX and the Linear kit has only just become available this afternoon.?

Theoretically: yes, it would improve the PA on the uBITX - a lot of care was spent on the board layout, keeping traces symmetric and leads as short as possible (look at the 1mm or so between the IRF510 body and the PCB, for example!). It has very flat gain from 2-30MHz and can even produce 10W on 4m band (70MHz) though gain is 8dB down by then... during testing this amplifier could not be made to exhibit any instability (oscillation). It also has a much larger heatsink than the stock uBITX amp, and the heatsink is grounded which also prevents feedback.?

BUT - the problems of the uBITX are not all in the driver and final PA stages. Earlier stages are overdriven and don't have flat gain with frequency. Spurs are generated but not filtered. Simply replacing the final two stages of uBITX only fixes some of the problems, not all.?

So yes, this amp would improve the PA... but at the end of the day it can only amplify what you give it: if what you give it is already unclean, it won't fix that.?

73 Hans G0UPL

From: [email protected] [mailto:[email protected]] On Behalf Of iz oos
Sent: Tuesday, October 9, 2018 11:27 AM
To: [email protected]
Subject: Re: [BITX20] QSX 10W HF Linear PA kit

?

Remember this commandment MadRadioModder: 'if you want even performance 3-30mhz you need to design that predriver section to be flat'

Il 09/ott/2018 17:56, "MadRadioModder" <madradiomodder@...> ha scritto:
>
> This kit won’t make the spur and bandpass problems of the uBITx go away.? Period.? Those problems occur in the early stages of the uBITx.? To the contrary it has the opportunity to make them worse since the gain of the amp is flat from 3-20 MHz? It may, however be a nice replacement for the final two stages in the uBITx if you want to maintain a fairly flat gain in those two stages.? That’s all.
>
> ?
>
> MRM
>
> ?
>
> ?
>
>
> --
>
> …_. _._

Hans,

I just ordered one of Your 10 W PA kits and quickly read over the assembly instruction.

Concerning the UBITX: the PA delivers 26 dB of gain, so for 10 W output the input power for the PA is +40 dBm - 26 dB = +14 dBm This power can be easily achieved with only one MMIC e.g. one of the GALI amps--- or a discrete verions (BS170)?

As You also offer LPF kits, the problems of the uBITX can be obsolete, but requires extra space for these extra boards.

One thing I m missing are the information about the P1dB pointz ( 40 dBm)? and IMD performance.?

Henning Weddig

DK5LV

First off, the uBitx as is does pretty well on 80/40/20 meters.
Power drops off on the stock uBitx as we go up to 15 and 10 meters.
There's 6 months worth of discussion in this forum about how best to?
get power up to where it should be on those high bands.
Tough to do, most attempts failed when modifying the existing uBitx power amp.
Grafting the QSX amp into the uBitx could mostly solve that.
Could also solve issues of power amp energy coupling back into the IF amps,
especially if the power amp is off in it's own box.

But the current signal levels in the IF amps and mixers create trouble with spurs
on those upper bands.? Not too much point in going to the QSX power amp unless
we clean that stuff up first.? And while we're at it, improve IMD and carrier leakage
in those IF amps and mixers as well.
?
If we assume we increase local oscillator CLK2 into the first mixer at D1,D2 up to 7dBm,
then the maximum signal into it from the 45mhz IF during transmit should be at least 10dB down
from that, or about -3dBm.? The mixer probably has a loss of around 6dB, and the 30mhz LPF
might lose another dB.? So if we replace everything from Q90 on out to the IRF510's with a
new power amp, that new amp will see a max of -10dBm, or 0.1mW going in.
This new QSX amp has 26 dB of gain for 10W out (where 10W out is +40dBm),
so needs 40dBm-26db = 14dBm going into the QSX amp (as Henning has said).
We need at least an additional 24 dB of gain to go from -10 dBm out of the 30mhz LPF
to +14dBm into the QSX amp.? That could easily be done with two MMIC stages,
and there are many such MMIC's capable of delivering 14dBm of power to the QSX amp.

Some extra gain in those two MMIC's would be good so we?can keep the signal through
the mixers even lower, and have some headroom for an adjustment such as RV1 of the uBitx.

Mouser 630-ADA-4743-TR1G

If you plan to bring the power up to 20W by boosting the supply voltage into the QSX amp,
then you will need another 3dB of power going in, so 14dBm+3dB=17dBm into the QSX amp.
And at 20W, you would want heavier transmit LPF's than the ones provided on the uBitx.

Also, it may be a good idea to have a 3dB attenuator between the MMIC's and the QSX amp
to keep it stable.

It will be interesting to see how clean we can make the uBitx as an exciter.
May need better transistors in the 45mhz stage, certainly want less gain
on the two transmit side IF amps.? Want 7dBm of drive from the local oscillators.
Perhaps the transformers in the mixers are not ideal, especially at 45mhz.
Raj has made considerable progress in sorting this some of this out.
?
I've got a QSX amp on order.?
Looking forward to playing with it.
But the uBitx has other issues to fix before we make use of the QSX amp
to increase power out on those upper bands.

Jerry, KE7ER

On Tue, Oct 9, 2018 at 07:54 AM, Henning Weddig wrote:

Hans,

I just ordered one of Your 10 W PA kits and quickly read over the assembly instruction.

Concerning the UBITX: the PA delivers 26 dB of gain, so for 10 W output the input power for the PA is +40 dBm - 26 dB = +14 dBm This power can be easily achieved with only one MMIC e.g. one of the GALI amps--- or a discrete verions (BS170)?

As You also offer LPF kits, the problems of the uBITX can be obsolete, but requires extra space for these extra boards.

One thing I m missing are the information about the P1dB pointz ( 40 dBm)? and IMD performance.?

Henning Weddig

DK5LV