Date   

Re: Amplifier chain design - best practices

Henning Weddig
 

Allison,

one more point: for the load always stay off the "knee" i.e. the so called ohmic range. Above this point the MOSFET acts like a voltage controlled (control voltage is the gate to source voltage) current source.

Example: for VGS = 5 V stay awav from the range of VDS = 0 to about 1 V. As easily be seen the "knee" gets worse for higher drain currents.

A similar condition happens for the RD16HFF1, having a larger ohmic region--  and in addition a limited maximum drain source voltage. The IRF 510 can do better as it can be operated at higher drain -source voltages.

The result: as the maximum drain -source voltage swing for the RF signal is limited the maximum output power can be only achieved through the output transformer and its transformation value.

There is a good explanation of this relationship in the proceedings of the "UKW-Tagung Weinheim" a few years ago by the author Prof. Dr Jochen Jirmann DB1NV.

  Henning Weddig

DK5LV 

Am 23.09.2018 um 17:21 schrieb ajparent1/KB1GMX:

Tim, IMD can be caused by two things, saturation at high power and
cross-over distortion near idle conditions. You missed one.  Anytime the output is not a uniform multiple of the input.
Or what is know ans non linear behavior.

Good example the HFE of 3904 at 10ma is 100, at 100ma its 40.  So is the 
waveform causes it to swing from 10 to 100ma the gain supplied is varying.
The base current required to get 10ma and 100 ma is not ten times higher
it is more like 25 times higher and the change is varying for intermediate
currents.  

For power MOSfets this is pronounced at low gate voltages as first we have
threshold where below that there is no conduction. as we go up from that point
the curve is markedly curved then straightens and at some high current bends
again.

For information look at the Ids VS Vds for Vgs curves.  From that graph we can get a
load line that reflects input voltage.  Old tube hands likely the curves looking like 
many power tubes.   It is on the page upper right corner.  Do not let pulsed be a bother
as in some cases those curves represent power well above what the device can do
sustained.  Needless to say its hard to run the drain to zero volts and remain linear. 
It also indicates a Gm of about 1S  IE 1Vgs change causes a 1A Drain current change.

Allison


Re: Bitx40 vs. uBitx

R. E. Klaus <reklaus@...>
 

Right now we are in a low time of sunspots which is part of the Solar Flux equation that effects our Ionosphere and allows the upper HF bands to work (20 through 10 meters) Until the Solar Cycle starts to raise again, 20 and up will be spotty in whether it's open or not.


uBitx Manager Question

Curt M.
 

Hello, I upgraded to the KD8CEC firmware last night and I really like it. I see that there is also the uBitx Manager software. 

Im guessing that I can backup my existing settings with this software?  I’ve been trying to get everything dialed in as far as the BFO and the calibration settings go and occasionally I really get things whacked out. It would be nice if the manager software allows me to back things up and then restore it to a good point should I mess things up. 

Is that that one of the uses of the Manager Software?  Also, do I use the Manager software with the radio powered up with the 12v power supply or only by having the USB cable attached like when you program the radio?

Thanks,

Curt


Re: Bitx40 vs. uBitx

Jack, W8TEE
 

Daniel:

I have both and I can't really say there's a big difference. One position is save the $50, buy the BITX40, and spend that extra money to upgrade to the best antenna you can buy. On the other hand, check the bands out when there's a major contest on and you find some activity on all bands. Sometimes, I think thousands of us listen to the band, don't hear anyone, so we say "the bands dead" and we give up. If we all got on and tried a few CQ's before giving up, I'll bet we'd find the only reason the band is dead is because we didn't try. So, do you really want to throw all of your eggs into the 40M basket? Eventually, those dead bands are going to open up and you'll want to be part of it. I just hope I'm around to see it!

Jack, W8TEE

On Sunday, September 23, 2018, 11:06:40 AM EDT, Daniel <gonewiththeego@...> wrote:


Hey everyone! I have a uBitx for few days and I can say that it's awesome! it's my first and only HF transceiver. Yesterday I listened all across 10m to 80m with just few meters of copper wire as an antenna. I can say that was... radio silence almost all over except for the 40m band. Now that I can see 40m band as a quite crowded place, I'm wondering... is there any advantage of having a Bitx40 comparing to uBitx? Is the reception or the transmission of the smaller brother somehow better comparing to the other one, giving the fact that Bitx40 is focused for 40m band only? Because now I'm wondering maybe uBitx wasn't the right choice for me and I should opt out for Bitx40. If we speak only about 40m band, can we say that Bitx40 is better than uBitx or not ? Thanks, and I strongly appreciate any kind of information and comment.


Re: Low Pass filter inductor question. #filters

davedt1e@...
 

Perfect!  Thanks Allison!  I'm doin it!


Re: Bitx40 vs. uBitx

davedt1e@...
 

Daniel,

I'm a bit of a noob, but I have been playing with with various transceivers in receive mode for a few years.  For what it's worth, I can tell you that regardless of the unit, your going to need more antenna to get decent reception.  Even if it's just taped to the ceiling where you live.  Up in a tree is better.  If it's for receiving only, you can use any cheap wire of any size.  Short antenna and bad band conditions = no reception.

Do this and I think you'll be pleased.


Re: Low Pass filter inductor question. #filters

ajparent1/KB1GMX <kb1gmx@...>
 

Air core has one liability that is mutual coupling.

So if you use air core the parts must be oriented so they will not couple
That means one in each plane of X, Y and Z.  Supporting the Z might be 
annoying (coil spirals away from the board).

Air core want to be no less than 1 coil diameter from any metallic surface
or wall. The rule of thumb for good Q is three coil diameters.

That and they will be bulky.  Wire size for 10W is anything you can
make mechanically stiff for self supporting and at least #28 for wound
on former.  Heavier gauge wire is handy for that.

For this close wound is likely ok.  Note higher Q may impact filter characteristics.
 
Allison


Re: Amplifier chain design - best practices

jim
 


The bother I have with most all those curves is that they start at Amperes ....Not mA's ...Been considering generating some on my own, just for fun and enjoyment ...(conductance curve probably)

Jim

On Sunday, September 23, 2018, 8:21:24 AM PDT, ajparent1/KB1GMX <kb1gmx@...> wrote:


Tim,


 

For information look at the Ids VS Vds for Vgs curves.  From that graph we can get a
load line that reflects input voltage.  Old tube hands likely the curves looking like 
many power tubes.   It is on the page upper right corner.  Do not let pulsed be a bother
as in some cases those curves represent power well above what the device can do
sustained.  Needless to say its hard to run the drain to zero volts and remain linear. 
It also indicates a Gm of about 1S  IE 1Vgs change causes a 1A Drain current change.

Allison


Low Pass filter inductor question. #filters

davedt1e@...
 

I'd like to build a few lpfs for my ubitx.  Can I use air core inductors?  I've got plenty of magnet wire and there are lots of inductor calculator online.

If this is doable,  Is close wound ok?  Also, what effect would wire size have?

Thanks in advance.


Re: Amplifier chain design - best practices

ajparent1/KB1GMX <kb1gmx@...>
 

Tim,
IMD can be caused by two things, saturation at high power and

cross-over distortion near idle conditions.

You missed one.  Anytime the output is not a uniform multiple of the input.
Or what is know ans non linear behavior.

Good example the HFE of 3904 at 10ma is 100, at 100ma its 40.  So is the 
waveform causes it to swing from 10 to 100ma the gain supplied is varying.
The base current required to get 10ma and 100 ma is not ten times higher
it is more like 25 times higher and the change is varying for intermediate
currents.  

For power MOSfets this is pronounced at low gate voltages as first we have
threshold where below that there is no conduction. as we go up from that point
the curve is markedly curved then straightens and at some high current bends
again.

For information look at the Ids VS Vds for Vgs curves.  From that graph we can get a
load line that reflects input voltage.  Old tube hands likely the curves looking like 
many power tubes.   It is on the page upper right corner.  Do not let pulsed be a bother
as in some cases those curves represent power well above what the device can do
sustained.  Needless to say its hard to run the drain to zero volts and remain linear. 
It also indicates a Gm of about 1S  IE 1Vgs change causes a 1A Drain current change.

Allison


Bitx40 vs. uBitx

YO8UFO
 

Hey everyone! I have a uBitx for few days and I can say that it's awesome! it's my first and only HF transceiver. Yesterday I listened all across 10m to 80m with just few meters of copper wire as an antenna. I can say that was... radio silence almost all over except for the 40m band. Now that I can see 40m band as a quite crowded place, I'm wondering... is there any advantage of having a Bitx40 comparing to uBitx? Is the reception or the transmission of the smaller brother somehow better comparing to the other one, giving the fact that Bitx40 is focused for 40m band only? Because now I'm wondering maybe uBitx wasn't the right choice for me and I should opt out for Bitx40. If we speak only about 40m band, can we say that Bitx40 is better than uBitx or not ? Thanks, and I strongly appreciate any kind of information and comment.


Re: Amplifier chain design best practices

ajparent1/KB1GMX <kb1gmx@...>
 

Jim,

Its not only if you do not meet "x" your problems start earlier.
Its the difference between ON and "well sorta on" the latter being conducting but not hard..
For commutating mixers (DBMs and other ) just sorta conducting is not on or off and
non-linearity is then the problem.

In the end not enough means exactly that.

Allison


Re: Amplifier chain design - best practices

Tim Gorman
 

IMD can be caused by two things, saturation at high power and
cross-over distortion near idle conditions.

I suspect the high standing current prevents cross-over distortion
which is why you get better IMD performance.

tim ab0wr

On Sun, 23 Sep 2018 14:53:53 +0400
"Ashhar Farhan" <farhanbox@...> wrote:

Allison,
Modelling confirmed to me that fewer active devices (and hence higher
gain per stage), the better is the IMD. From an IMD perspective,
higher gain with high standing current results in better IMD.
That's why it makes sense to use a VHF power device as
driver/pre-driver block. I did a low distortion (less than - 40dbc
IMD) linear chain a few years ago. At 12v, It had to be an RD16HHF1
device biases for close to class A. At 1A of current, it gave 4 watts
of power. The details are on the emrfd group's file section.
- f


Re: Off-topic posts, Amplifier chain design

Jack, W8TEE
 

Mike:

I think your criteria may be too narrow. Not all of us are EE guys who understand the nuances of amplifier design. To me, most of the topics you're complaining about seem to make sense to me, perhaps because I'm not an EE person. Even something that appears off topic (e.g., K2 and its price) make sense to me since there needs to be a standard against which things are measured. Personally, judging against a well-known product is more insightful than fitting data to an equation from a book.

Often I read posts that I know I probably won't completely understand (like amplifier chain design) because I hope to glean some information from that post. Given that there are thousands of us here and a good chunk are not engineers, "off topic" to some may actually be the post that makes "the penny drop" for others.

Finally, we have a very capable monitor for this group who is unafraid to jump in when he feels the nature of a post or its thread doesn't serve the group's interests. I suggest that we leave the policing of this group to his judgement.

Jack, W8TEE


On Sunday, September 23, 2018, 12:25:42 AM EDT, K5ESS <k5ess.nothdurft@...> wrote:


If there was ever an appropriate comparison, the now overused phrase “like herding cats” fits the bill with regard to keeping these group posts on topic.

Mike

K5ESS

 

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of ajparent1/KB1GMX
Sent: Saturday, September 22, 2018 2:21 PM
To: BITX20@groups.io
Subject: Re: [BITX20] Amplifier chain design - best practices

 

Did a count and nearly 50% of the posts in thsi string have nothing to do with:

amplifier chain design - best practices.

About DBMs,
DBM drive levels
Low pass filters
Spectrum analyzer and probes.
K2 and its price.
SSD
QUCS

I might have forgotten a few.

Someone change the title to "WHATEVER/"


Re: Ubitx evolving fixes updated to wiki? #ubitx #ubitx-help

davedt1e@...
 

Thanks folks!  I found some good pages on filters to study up on.  Doesn't look too bad.  Think I'll go ahead and build them for the experience points :)


Re: Amplifier chain design - best practices

ajparent1/KB1GMX <kb1gmx@...>
 

Farhan,

I am very aware that fewer devices are desired for better IMD.   The key is using
competent devices at attainable gains.  There are two problems with 3904s in that
role, gain compression at higher operating currents and a distinct problem with
delivering the desired gain at higher frequencies.   If they cannot produce the
needed gain any feedback required for supporting input and output impedance
will be unpredictable producing a cascade of performance degradation.  Its not
what you may have built in some prior time as we have many tested that do not.   
Its about how to do it at some production level.

First is we do not ask a device to push above it weight class.  2n3904 is not going
to deliver more than 12DB power gain at 3-30mhz and less than 1db gain flatness.
The power it can produce is not device dissipation its gain at currents higher than 
its peak current gain (about 10mA for HFE peak and HFE is well down at 100mA).
This shows well of a curve tracer.  We would deire for linear use a prt that has a
more or less flat current gain in the operating range.

By measurement we have insufficient drive and gain at frequencies approaching 8mhz.
As a result of those factors the IMD measured is seriously poor above 2W even on bands
where more power can be delivered.  The key here is power with acceptable IMD from
3 to 30mhz not 3 to maybe 10mhz.   We cannot do that with devices that incapable of
that for those first three stages. 

Note this is not an IRF510 issue as the driver has failed to deliver adequate drive and
the prior stages requires overdrive to function at all.  That does not help IMD or spurs.

Also bias of 1A for RD16HHF is not class A above about 3 to maybe 4W and wanders
into AB1 above that power.   A pair of IRF510s at 12V can achieve that at about
250ma though at higher voltages can easily surpass that.  In general the RD16HHF is
about $4.50 more expensive to get maybe 3db better IMD at 12V if everything
around it supports that.  To do that prior stages have to built to do that.  The IK4AUY
amp (IRF510 pp driving IRF510PP) supports the ability of the part to perform.

In either case loading a IRF510 or RD15HHF to 50 ohms Drain to Drain is far too high
a load resistance for more than a few watts.  Reason is MOSFETs are voltage driven
current sources.   If you want power from them the load current must swing in the
linear current region without exceeding the available voltage.  So with 50 ohms and
not more voltage say 12V peak 2.88W is about it and we are driving devices to
saturation.   We need s a load line what can support a Drain current in the 1-2A region.
for that was need to get to 12.5ohms for a 10W amp without saturating or voltage clipping.
At that point we are running near 1A Peak current and likely with 200-500ma of bias we
are in the linear region for a push pull amp.  Combined with a reasonable feedback for
flatter gain vs frequency the IRF and RD are in the same region.   But this will be true
only if the drivers and prior stages are not suspect.

Allison


Re: Amplifier chain design - best practices

Ashhar Farhan
 

Allison,
Modelling confirmed to me that fewer active devices (and hence higher gain per stage), the better is the IMD. From an IMD perspective, higher gain with high standing current results in better IMD.
That's why it makes sense to use a VHF power device as driver/pre-driver block. I did a low distortion (less than - 40dbc IMD) linear chain a few years ago. At 12v, It had to be an RD16HHF1 device biases for close to class A. At 1A of current, it gave 4 watts of power. The details are on the emrfd group's file section. 
- f

On Fri, 21 Sep 2018, 21:21 ajparent1/KB1GMX, <kb1gmx@...> wrote:
On Fri, Sep 21, 2018 at 09:19 AM, Jeff Omundson wrote:
1)Keep amplifier intos and outofs away from each other.

This also applies to all RF systems.  

2)Keep amplifier gain stages reasonable.  Add more stages instead of upping gain.

This is a serious and common error, forcing or wishing for gain that is not predictably there.
Conservative design rules insure that.

3)Pick the right device for the desired frequencies.  Obviously 2n3904's don't work for this application. 

First the 2n3904 can if one does not apply wishful thinking and employs it within the devices capability.

***Does picking a part with an fT about 40 times the highest desired frequency a safe bet? 
Oddly the answer is yes if the stages has to deliver at least part of the gain desired.  For a 
30mhz amp with a 20db stage gain you will need an Ft of something approaching 3000mhz!
The real question is does it makes sense when two 3904s at 10DB each can also do that 
at 30mhz.  

***Can you pick an FT that is too high (like a npn transistor with and fT of 10 gHz for a 3-30mHz radio)?  
The answer is yes and maybe.  IF you using resistive feedback amps (nearly the entire transmitter and both IFs)
the required FT is a product of the gain and the frequency so at some point you really have to ask is that much
gain per stage reasonable and stable.  IT might be prudent to back down the gain and use a more reasonable
transistor.  Also for higher power stages the cost for a supergain high power device might break the bank.

A lot of QRP designs more so older one try to be minimal and get the most out of a very few parts.
If you making one this is ok as you can dial it in and hand pick parts.  For quantity builds this is terrible.
You want every one to be the same as the last.  So care in not trying to get 101% of capability is advised..
This shows more often in transmitter circuits as out of a group most doe the expected watts, a few do
more and a bunch never do that.  Multiband transmitters due to bandwidth, power and layout require
greater attention.

4)Keep RF wires/traces away from each other and from the dc path.
Again this works for RX and TX circuits.  We do not want o induce unwanted signals into the receiver.

From Crystal Sets to Sideband, a very good starting place.  My Favorite are Solid State Design For Radio Amateurs,
and Experimental Methods in RF Design (Both ARRL).  Neither are light bits and do require several reading through
passes to pick up that what was mentioned here is often elaborated and related to things there.

***Makes me wonder how the ft817 does hf, vhf and uhf... 

Lots of tuned filters, careful layout the use of the chassis as cooling, shielding, and mechanical support.

Its worth looking at Elecraft K1 and K2 as the manuals and detail in them is very good as examples.
I'd add KNQ7A  and the KD1JV Sopbucket 20 as a simple monoband, and a few other as well.  Generally
the designers are creative people so watch for solutions to problems not outwardly discussed as its
often not featured but important.  Cooling and layout are often not obvious without analysis.  Same can
be said for component choices.


Allison


Re: Amplifier chain design - best practices

K5ESS
 

If there was ever an appropriate comparison, the now overused phrase “like herding cats” fits the bill with regard to keeping these group posts on topic.

Mike

K5ESS

 

From: BITX20@groups.io [mailto:BITX20@groups.io] On Behalf Of ajparent1/KB1GMX
Sent: Saturday, September 22, 2018 2:21 PM
To: BITX20@groups.io
Subject: Re: [BITX20] Amplifier chain design - best practices

 

Did a count and nearly 50% of the posts in thsi string have nothing to do with:

amplifier chain design - best practices.

About DBMs,
DBM drive levels
Low pass filters
Spectrum analyzer and probes.
K2 and its price.
SSD
QUCS

I might have forgotten a few.

Someone change the title to "WHATEVER/"


Re: Amplifier chain design best practices

jim
 

Maybe not written stone, but rather silicon ...All diodes need a minimum threshold voltage to "turn on" ie start conducting ...Silicon happens to be about .7 volts, schottky about .3 volts ...Level "7" mixers have one diode per leg in the bridge,  Level "14" have 2 diodes in series per leg ..etc.   You can build your own out of 1n914's if you wish ...Still gonna take X amount of volts to turn em on...If you don't meet "x" then your problems start ..

Jim

On Saturday, September 22, 2018, 4:46:11 PM PDT, Timothy Fidler <engstr@...> wrote:

st

Amplifier chain design - best practices  WHATEVER.

For whom it may concern : The 7 dBm/50Z  figure comes from a requirement which was common for the Minicircuits series of  DBMs  viz the ADE-01 and SBL-1 which used to be very common in ARRL Imprimatur designs.  It is not written in stone.  Some other  Minicircuits Dbms require other input power to meet the mafr's datasheets. 


Re: Ubitx evolving fixes updated to wiki? #ubitx #ubitx-help

Mark M
 

Low pass filter kits can be had from QRPLabs or you can build your own. The QRPLabs instruction documents have all the values they use...it would be easy to roll your own on a piece of perf board. Or some of the members here offer circuit boards. The components are not expensive or hard to find (altho it might be more cost effective to order them from QRPLabs). Lots of info online or in the ARRL Handbook on design.

Or see the wiki for details on how to hack the existing filters on the uBitX (thanks to Allison, et al.).

See https://www.qrp-labs.com/lpfkit.html for more info on the kits (disclaimer...I have no financial interest in QRPLAbs, just a customer).

73... Mark AA7TA

On 9/22/18 7:41 PM, ajparent1/KB1GMX wrote:
...
Spurs are less an issue for 20M and below.
External low pass or transmitting band pass filters are responsible way to go.