(tr)uSDX power, efficiency and filter optimisation

Post Reply
Faraaz VK4JJ
 

Posts: 215
Joined: 30 Dec 2021, 21:56

Re: (tr)uSDX power, efficiency and filter optimisation

Post by Faraaz VK4JJ »

Here are some measurements to support the theoretical calculations.

7 turns, evenly spaced, of 0.42mm copper enamelled wire, well-scraped and leads tinned. Leads are about 4mm long. Measured on a jig, OSL calibrated 13-15MHz.

Z= 541 +j465
Ls = 5.3uH

If we assume 50mA, this would indicate 1.3W resistive losses at 14mhz!

Image
W6LKB
 

Posts: 15
Joined: 31 Dec 2021, 14:30

Re: (tr)uSDX power, efficiency and filter optimisation

Post by W6LKB »

hgosch1 wrote: 22 Mar 2022, 16:29 Hello friends of the famous truSDX!

I did it my way….
My fully assembled truSDX worked right ok, but the transmission power was only 2.5 W on all bands at 12V.3 V and a current consumption of only approx. 370 mA.

Since I had already successfully used broadband transformers in receiving systems, I built a 1:2 transformer (4+4 windings) with a double hole core as a measuring aid. This enables a resistance transformation of 1:4 from the RF point to the port of the NanoVNA, making it possible to evaluate both filter parts in a 50 ohm system.

This allows the adjustment of Lx1 and LX2 to be displayed quite comfortably in the 50 ohm system of the NanoVNA and is used for the filter adjustment too.

With this, I was able to very fast achieve output powers of 5 to 6.84 watts with efficiencies between 71 and 78% at 12.3V Ub right away on 80 to 30m.

The measurement itself takes place, as suggested by Manuel, with the only difference that the start frequency is now set slightly below the respective band frequency and the stop frequency is just slightly above the frequency of the first harmonic. So for example 3-30 MHz.

Two dips can be seen in the S11 curve at NanoVNA-saver (Displaying S11 and S21). For 30m filter adjustment the first dip here was at about 8MHz and is determined by L21.

In the S21 curve we see the LPF curve and a dip at just around 20 MHz.

Now it is clear at a glance what needs to be done.
1. Bring L21 to lower inductance to shift the dip to the 30m band.
2. Bring L22 to a inductance in order to shift the S21 dip as precisely as possible to 2x the transmission frequency.

With my truSDX it was necessary to reduce L21 by 3 turns. L22 could be brought to exactly twice the operating frequency with a larger spread of the windings.

Without further activities, the changes to L21 and L22 resulted in an a good increase in the TX current consumption at 12.3V and the TX output power from approx. 2.5 watts to 5.22 W with an efficiency of 76% shown on the OLED display.

Bingo! This method also made it possible to achieve similar values on the other bands! In most cases 2-3 turns had to be taken off from Lx1 and one turn from Lx2.

So far I've only been able to get 3.15 watts at 50% efficiency with this method on the 20m band, so more troubleshooting is needed. First I will probably remove the two inductors and check the values of the capacitors fitted.

Does anyone have another idea to narrow down my 20m problem?

73, Harry oe6gc
Can you explain how you built the transformer and where you connected it exactly?

Thank you!
hb9hem
 

Posts: 36
Joined: 07 Jan 2022, 04:39

Re: (tr)uSDX power, efficiency and filter optimisation

Post by hb9hem »

I think you just create a 1:2 autotransformer (8 windings total, pick primary off the 4th winding), hook up the low impedance side to the board and the high impedance side to the VNA. That transforms the impedance 1:4. Maybe 1:3 in impedance would be better, i.e. a 4:7 transformer?
g7jur
 

Posts: 35
Joined: 18 Feb 2022, 22:49

Re: (tr)uSDX power, efficiency and filter optimisation

Post by g7jur »

This is the results of mine, at 13.8 Volts.
*Rxxxx* kit.

80M 4.84W 70.29%
60M 5.29W 71.49%
40M 3.73W 69.26%
30M 4.85W 69.29%
20M 3.64W 61.69%

I have changed the 3 FT37-43, and all the BS170's

Its a bit better.

80M 6.58W 72.31%
60M 5.54W 74.00%
40M 3.93W 70.00%
30M 5.54W 71.15%
20M 4.09W 63.91%

Philip G7JUR
Last edited by g7jur on 01 Apr 2022, 21:38, edited 1 time in total.
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: (tr)uSDX power, efficiency and filter optimisation

Post by DL2MAN »

Faraaz VK4JJ wrote: 28 Mar 2022, 23:13 Here are some measurements to support the theoretical calculations.

7 turns, evenly spaced, of 0.42mm copper enamelled wire, well-scraped and leads tinned. Leads are about 4mm long. Measured on a jig, OSL calibrated 13-15MHz.

Z= 541 +j465
Ls = 5.3uH

If we assume 50mA, this would indicate 1.3W resistive losses at 14mhz!

Image
While this is absolutely plausible, one question still remains:
Why do we get 80% Plus Efficiency on Chinese Kit and our own Test Kits, but not with the *Rxxxx* Kit ?They are using the same high quality Ferrites (from Fair Rite) as we do (Amidon sells actually FairRite) There must me a critical element involved, that causes extra 10-20% Loss and put it out of specs.
But I can´t find it.... My other cores should have arrived today, but they are stuck.... Hope they arrive tomorrow.

73 Manuel; DL2MAN
g7jur
 

Posts: 35
Joined: 18 Feb 2022, 22:49

Re: (tr)uSDX power, efficiency and filter optimisation

Post by g7jur »

DL2MAN wrote: 31 Mar 2022, 16:09
Faraaz VK4JJ wrote: 28 Mar 2022, 23:13 Here are some measurements to support the theoretical calculations.

7 turns, evenly spaced, of 0.42mm copper enamelled wire, well-scraped and leads tinned. Leads are about 4mm long. Measured on a jig, OSL calibrated 13-15MHz.

Z= 541 +j465
Ls = 5.3uH

If we assume 50mA, this would indicate 1.3W resistive losses at 14mhz!

Image
While this is absolutely plausible, one question still remains:
Why do we get 80% Plus Efficiency on Chinese Kit and our own Test Kits, but not with the *Rxxxx* Kit ?They are using the same high quality Ferrites (from Fair Rite) as we do (Amidon sells actually FairRite) There must me a critical element involved, that causes extra 10-20% Loss and put it out of specs.
But I can´t find it.... My other cores should have arrived today, but they are stuck.... Hope they arrive tomorrow.

73 Manuel; DL2MAN
I did notice that T2 gets warm.
Philip G7JUR
hb9hem
 

Posts: 36
Joined: 07 Jan 2022, 04:39

Re: (tr)uSDX power, efficiency and filter optimisation

Post by hb9hem »

I made some measurements. I don't know much about class E, but I do know that the voltage at RF is not what it should be.

Yellow: ANT_OUT
Blue: DRV
Pink: RF
Image
Image
Ohwenzelph
 

Posts: 207
Joined: 01 Jan 2022, 03:47

Re: (tr)uSDX power, efficiency and filter optimisation

Post by Ohwenzelph »

Sometimes it's not the ferrites, sometimes it's the caps.
AC9YY
 

Posts: 133
Joined: 21 Mar 2022, 00:04
Location: Illinois, USA

Re: (tr)uSDX power, efficiency and filter optimisation

Post by AC9YY »

RF is pre-filter, so I think that doesn't look too bad.
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: (tr)uSDX power, efficiency and filter optimisation

Post by DL2MAN »

hb9hem wrote: 31 Mar 2022, 20:33 I made some measurements. I don't know much about class E, but I do know that the voltage at RF is not what it should be.

Yellow: ANT_OUT
Blue: DRV
Pink: RF
Image
Image
Cool, a 4 channel scope.... I only have 2 channels....
However, your "RF" looks OK. RF is connected to drain, and it´s a textbook example of a class E Waveform.
What I can see from your curve is actually, that your switching times are suboptimal, as there is still energy in the system, when the fet is already switched on again by "drive"
Look at the original Sokal Paper. He has examples of those waveforms and how to optimize them.
https://people.physics.anu.edu.au/~dxt1 ... design.pdf
(Page 8 of the Document, Page 16 according to Index)

73 Manuel; DL2MAN
S53SL
 

Posts: 9
Joined: 31 Mar 2022, 21:24

Re: (tr)uSDX power, efficiency and filter optimisation

Post by S53SL »

Hi folks.
I've built my Up Tech Store AliExpress kit, and tested it. But I'm getting out low power, mostly bellow 2W and efficiency 40% or worse. SWR on dummy load is almost 1:1

I'm not sure where I can begin solving a problem. Windings on coils are equally spaced and squeezing doesn't do much difference. Is it possible there is something wrong with one of the fets?

My measurements:
80m: 1.2W 25%
60m: 1.8W 40%
40m: 2.8W 40%
30m: 1.5W 32%
20m: 1.9W 41%
Attachments
IMG_20220331_233208a.jpg
IMG_20220331_233208a.jpg (185.22 KiB) Viewed 6285 times
73 de S53SL
g7jur
 

Posts: 35
Joined: 18 Feb 2022, 22:49

Re: (tr)uSDX power, efficiency and filter optimisation

Post by g7jur »

S53SL wrote: 31 Mar 2022, 21:42 Hi folks.
I've built my Up Tech Store AliExpress kit, and tested it. But I'm getting out low power, mostly bellow 2W and efficiency 40% or worse. SWR on dummy load is almost 1:1

I'm not sure where I can begin solving a problem. Windings on coils are equally spaced and squeezing doesn't do much difference. Is it possible there is something wrong with one of the fets?

My measurements:
80m: 1.2W 25%
60m: 1.8W 40%
40m: 2.8W 40%
30m: 1.5W 32%
20m: 1.9W 41%
Hello
This is no good, I wish I could help you.
There are same problems with the kits, that need to be resolved.
Philip G7JUR
John
 

Posts: 49
Joined: 19 Feb 2022, 08:56

Re: (tr)uSDX power, efficiency and filter optimisation

Post by John »

S53SL wrote: 31 Mar 2022, 21:42 Hi folks.
I've built my Up Tech Store AliExpress kit, and tested it. But I'm getting out low power, mostly bellow 2W and efficiency 40% or worse. SWR on dummy load is almost 1:1

I'm not sure where I can begin solving a problem. Windings on coils are equally spaced and squeezing doesn't do much difference. Is it possible there is something wrong with one of the fets?

My measurements:
80m: 1.2W 25%
60m: 1.8W 40%
40m: 2.8W 40%
30m: 1.5W 32%
20m: 1.9W 41%
Off subject, but it looks to me like you have 8 turns on T1. The magic number is 7.
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: (tr)uSDX power, efficiency and filter optimisation

Post by DL2MAN »

John wrote: 01 Apr 2022, 02:27

Off subject, but it looks to me like you have 8 turns on T1. The magic number is 7.
That´s not off topic, that´s a big fat BINGO !
hb9hem
 

Posts: 36
Joined: 07 Jan 2022, 04:39

Re: (tr)uSDX power, efficiency and filter optimisation

Post by hb9hem »

However, your "RF" looks OK.
Actually, I came to a different conclusion. According to Fig.4 on p.15 of Sokal's exposition this "RF" voltage is a "typical mistuned V_CE waveform". The trough must be moved to the FET ON switching time so that "RF" looks like in Fig.3. The remedy, according to Fig.5, is to increase the resonator L and C. Does anyone have an intuition about how much increase is necessary? (I have to economize because I don't have many C's lying around to play with.)
M1KTA
 

Posts: 26
Joined: 17 Jan 2022, 18:05

Re: (tr)uSDX power, efficiency and filter optimisation

Post by M1KTA »

DL2MAN wrote: 17 Mar 2022, 16:14 Could you please specify
- which batch you have (1st or 2nd) ?
- what bands are affected by heated caps ?

I am in direct contact with them and we're trying to figure out what's wrong.

73 Manuel DL2MAN
Manuel,

Mine is Batch 2 from *Rxxxx* so almost certainly one of those effected.

I have some older FT37-43 and T37-2 from kits and parts from a few years back and they are known good, as well as some BS170 from KD1JV so I expect I will look to remake the RF board. I was thinking of doing a WARC and a Classic one anyway as it is kind of modular... I guess might look to one of the group buys and see if anyone getting some PCB made up and might have a spare or two? (Happy for contacts if anyone is BTW).

The C I'll look and see what I have but might have to order the 'correct' types (C0G/NPO 100V and not X7R 50V as were fitted).

73
Dom
M1KTA
DL2MAN
 

Posts: 706
Joined: 30 Dec 2021, 19:18
Contact:

Re: (tr)uSDX power, efficiency and filter optimisation

Post by DL2MAN »

Hey guys,

Today, the new Ferrite-Toroids came. I swapped all 3 of them on the *Rxxxx* Board and it did not change anything.
I have now exchanged everything on that Board in the RF Section, and still cannot get above 65% Efficiency. I will no longer waste my time with this. I´ve spend now a couple of hours and some € on trying to fix it. You gotta know when it´s over. Sorry guys.
Even though it hurts my ego, I will give up, as my spare time is rare and precious. I hope someone figures out what is wrong on *Rxxxx* Boards. :-/

73 Manuel; DL2MAN
Iw2fpe
 

Posts: 12
Joined: 20 Mar 2022, 17:06

Re: (tr)uSDX power, efficiency and filter optimisation

Post by Iw2fpe »

Hi Manuel,
Do you confirm that you have tried a non-*Rxxxx* RF card on *Rxxxx* mainboard and have obtained the correct efficiency?
Just to point the research only on the RF part, at the moment I have replaced all the toroids, I have to solve a SWR problem, and then I try with the capacitors.

73
Lino, Iw2fpe
g7jur
 

Posts: 35
Joined: 18 Feb 2022, 22:49

Re: (tr)uSDX power, efficiency and filter optimisation

Post by g7jur »

DL2MAN wrote: 01 Apr 2022, 19:26 Hey guys,

Today, the new Ferrite-Toroids came. I swapped all 3 of them on the *Rxxxx* Board and it did not change anything.
I have now exchanged everything on that Board in the RF Section, and still cannot get above 65% Efficiency. I will no longer waste my time with this. I´ve spend now a couple of hours and some € on trying to fix it. You gotta know when it´s over. Sorry guys.
Even though it hurts my ego, I will give up, as my spare time is rare and precious. I hope someone figures out what is wrong on *Rxxxx* Boards. :-/

73 Manuel; DL2MAN
I have changed the 3 FT37-43, and all the BS170's

Its a bit better.

80M 6.58W 72.31%
60M 5.54W 74.00%
40M 3.93W 70.00%
30M 5.54W 71.15%
20M 4.09W 63.91%

T2 still getting hot why ?. That's were the power is getting lost.

Philip G7JUR
g7jur
 

Posts: 35
Joined: 18 Feb 2022, 22:49

Re: (tr)uSDX power, efficiency and filter optimisation

Post by g7jur »

I would like to get it working properly
Last edited by g7jur on 02 Apr 2022, 10:03, edited 1 time in total.
6k2kus
 

Posts: 8
Joined: 19 Mar 2022, 02:14

Re: (tr)uSDX power, efficiency and filter optimisation

Post by 6k2kus »

What about this possibility? Maybe the kit works fine, but the efficiency measurement may be wrong.

I made the power around 5W, mostly by unturning the coil once. I use external wattmeter with dummy, so I can confirm that the displayed power is true.

The efficiency is less than 50% on every band. But I doubt the displayed efficienty of mine. What is strange is the transistors do not heat up. I understand that this is a signal of high efficiency, because when I set up wrongly before, they heated up quickly with lower efficiency. This is easy to see if you touch the transistors. That time I could not touch them more than one second.

To verify this, I pushed the PTT about 16 seconds, which is a complete cycle of FT8. It almost did not heat up. Also there is less spurious radiation than -43dB. From the first law of thermodynamics, the inefficient energy should go somewhere and heat up.

If this is so, I am willing to use it under this condition even if I fail to achieve as high efficiency as 80%.

Then my question now is, how can we verify that the displayed efficienty is correct? I imagine the straightforward is to use galvanometer in the input. Maybe I constructed the measurement part wrongly. Any suggestions?
hb9hem
 

Posts: 36
Joined: 07 Jan 2022, 04:39

Re: (tr)uSDX power, efficiency and filter optimisation

Post by hb9hem »

Maybe the kit works fine, but the efficiency measurement may be wrong.
Have a look at this post. The efficiency truly is too low.

Personally, I have two reasons to pursue this. Firstly, going from 60% to 80% efficiency gives me 25% longer battery life. Secondly, I find the whole design quite nifty and I just want it to reach its potential. But, tbh, my interest lies purely in the class E amplifier. The digital part is closed-source, so I don't care much about it.
You should stop *Rxxxx* selling sub substandard kit's
Are they really knowingly selling "bad" kits? If yes, that would mean gloves off for me.
ea5anx
 

Posts: 5
Joined: 01 Jan 2022, 10:41

Re: (tr)uSDX power, efficiency and filter optimisation

Post by ea5anx »

*Rxxxx* (tr)uSDX mainboard is tested to be ok by Manuel.
Has someone mounted RF board *Rxxxx* parts on a non *Rxxxx* RF board?
If this way it had a good efficiency, supplying new RF boards, SMT mounted, would not be too expensive but time saving.
5 RF boards about 15€ plus handling, 50 RF boards 7€ + plus handling.
SA4DHT
 

Posts: 19
Joined: 10 Jan 2022, 14:13

Re: (tr)uSDX power, efficiency and filter optimisation

Post by SA4DHT »

I have a kit from *Rxxxx* batch 3. I am facing the same issues regarding efficiency as others have reported at 12 V in:
20 m 3.8W 45.7%
30 m 3.4W 50.0%
40 m 2.7W 50.5%
60 m 2.6W 51.5%
80 m 3.8W 51.1%

Started to check 20 m with a high resolution thermal camera and find C12, C19 and C16 getting very hot about 70 deg C. So there energy is wasted. I will try to replace them. I have NP0 here but not sure if they are 50V or 100V. Another question is how the target capacitance is achieved on good versus bad board. Wrong type and voltage is one possibility another if one or two capacitors are used per position in order to get the goal capacitance. MOSFET BS170 are getting hot too at about 40 deg C one beeing at higher temperature so current sharing is not perfect. T2 is getting hot too about 40 deg C. On one of the videos Manuel has posted it seems to me T1 and T2 are winded with bigger diameter wire than 0.4 mm.
SA4DHT
 

Posts: 19
Joined: 10 Jan 2022, 14:13

Re: (tr)uSDX power, efficiency and filter optimisation

Post by SA4DHT »

There is a mismatch between schematic/BOM and the table with L and C values for the different transistors and frequencies. For 20 m schematic/BOM for C10//C11 is 68 pF in the table for 3 x BS170 it says 128 pF for 20 m and CX0//CX1.
Post Reply

Users browsing this forum: Bing [Bot] and 1 guest