HL2 Build 8 Observations

[John Williams]

I now have my board up and running.

Observations so far -

My board thickness was 1.8mm. Issue resolved, this is just for awareness.

My PA transformer had a wire loose.

It is re-soldered and working. This was one side of the single loop pair of wires.

I then tried various power settings but noticed that I was not getting full 5W power output. I checked the bias using quisk and noted that it was pulling 130ma with spot to 0 as is described in Steve's excellent video. I re-calibrated the bias to a full 200ma. I don't know if perhaps the transformer was not connected during the testing and still somewhat worked or what. So I suggest you check your bias levels after you get things going.

My SMA connectors were undamaged.

So far, so good. Happy camper!

John - W9JSW

[in3otd]

Thanks John,
I would have expected the loose wire to be detected during the final test, since the bias for one of the LDMOS would have been always at zero - but I did not check exactly how the test code works.

My boards measure both 1.6 mm, maybe they come from a different batch?

I've measured the TX output power of my H-Lv2b8 with the filters in bypass and it's not as high as on the H-Lv2b3 I have: the long transformer wires cause some drop at the high end but even after shortening the wires the output is still lower than expected - see graph below which compares the build8 as received, build8 after shortening the T3 wires and the build 3 output:

the low-power output from the driver looks fine, almost identical to the build3. I'll try to find out where the difference comes from in the next days.

Are you able to get 5 W also on 10 m?

73 de Claudio, DK1CG / IN3OTD

[John Williams]

Quisk says I am getting 3.5W out. My meter (that I do not think is accurate) says 4W out - both on 40M. This is with the filters inline. I will run a few more measurements.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/2f2f2c31-303d-4be4-8071-fe710601b774%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Steve Haynal]

Hi John and Claudio,

Your broken wire and unprogrammed bias is a bit alarming. The test checks for bias current and should have failed. Your picture did not come through. At which end was the black wire broken? I will do the same here and verify the test behavior.

On one build 8 unit I checked power output on all bands. Everything was as expected. Please let me know if you learn anything more about your low power output.

On Quisk and SparkSDR, there is a procedure to calibrate the HL2 forward/reverse power readings. This should help with accuracy.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[Steve Haynal]

Hi Claudio,

So shortening the T3 leads boosts power on the higher frequencies, but not by as much as expected. The one unit I tested was the first unit I received. It had short leads and another type of wire with insulation that melted easily. I asked them to use better insulation and recommended the last wire we used, but they substituted something else. I wonder if the wire type accounts for the differences you still see. 

I will try the same experiment with one of the spare HL2s I have here with long leads.

73,

Steve

kf7o

[Bob A. Booey]

At the risk of stating the obvious, these types of Xfmr’s usually are constructed using Teflon insulated wire.

Sent from my iPhone

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/ded9e9e4-cf26-45cb-9619-acac44e9dff0%40googlegroups.com.

[Duncan Clark]

In message <3e2e2895-da3d-9391...@gmail.com>, John
Williams writes

>I then tried various power settings but noticed that I was not getting
>full 5W power output. I checked the bias using quisk and noted that it
>was pulling 130ma with spot to 0 as is described in Steve's excellent
>video. I re-calibrated the bias to a full 200ma. I don't know if
>perhaps the transformer was not connected during the testing and still
>somewhat worked or what. So I suggest you check your bias levels after
>you get things going.

Just tried Tx for first time using both PowerSDR (using Tune) and
SparkSDR (using FM) and forgot to check the initial bias current shown.

Plus I don't trust my power meter readings at low power

Anyway I have a real funny.

Running into HF quad or dummy load I am way down on power on 12m.
Max drive is only showing <1w and a maximum of 0.72A on SparkSDR. No
improvement with what I assume is filters out using an unpopulated J16
table in SparkSDR as opposed to the N2ADR LPF-HPF selection. So I think
that rules out a dodgy 10/12m LPF.

10m is showing 3W maximum at 0.94A. Get roughly same power showing on
17m and 15m. Lower bands into dummy load show more power, 5W or so at
max SparkSDR drive.

Could be the length of leads on that transformer having some odd effect
on 12m.

I haven't tried Quisk yet but will tomorrow and will also check the bias
setting tomorrow plus check current drawn on each band.

Duncan
--
Duncan Clark
G4ELJ

[Steve Haynal]

Hi Group,

I checked one of the Hl2 build8 units here with short leads, and the power output was low as reported on 15M,12M and 10M. The lowest was 12M at ~3.9W. Rewinding the balun with the wire used in the last run and keeping the leads very short did have some improvement, but power on the high frequencies was still not what I've seen in the past, maybe 4.4W on 12M. When I use one of my older N2ADR filter boards, the power is what is expected, just under 5W on 12M. It appears that a different capacitor manufacturer was used for C25, C39 and C43. I specified multiple acceptable part numbers this time around, so they are probably one from my approved list. Tomorrow I shall try to replace them on one N2ADR filter board to see if they are degrading power at high frequencies.

73,

Steve

kf7o

[John Williams]

Group,

I have only tested 40M with my radio using my antenna. I will plug in the dummy load and measure all of the bands this weekend.

For 40, quisk and SparkSDR show 3.4 watts and a close to 1.03 SWR. My meter somewhat agrees showing a bit under 4 watts. As I have stated before, I do not trust my second hand meter.

What should this radio be capable of on a per-band basis?

What wire should I use if I choose to rewind the transformer? Will it matter? All I have is enameled wire.

John

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/6845d7e8-2d57-4986-b88a-85f9e17fd55f%40googlegroups.com.

[in3otd]

Hello,

> What should this radio be capable of on a per-band basis?

from the measurements on the Wiki, when using the N2ADR filter board the output power should be around:

160 m    37.5 dBm    5.6 W
 80 m    37.0 dBm    5.0 W
 60 m    37.2 dBm    5.3 W
 40 m    36.8 dBm    4.7 W
 30 m    37.7 dBm    5.9 W
 20 m    37.6 dBm    5.8 W
 17 m    37.2 dBm    5.2 W
 15 m    36.6 dBm    4.5 W
 12 m    36.7 dBm    4.6 W
 10 m    36.7 dBm    4.7 W

I was under the impression the numbers were actually a little higher...

> What wire should I use if I choose to rewind the transformer? Will it matter? All I have is enameled wire.

the wire should not be critical but Teflon/Kynar insulation was recommended because the transformer core is conductive and has sharp edges so a tough insulator will avoid that the wires comes accidentally in contact with the core. I think also someone reported that the transformer could get quite hot when transmitting into a mismatched load.
Having said that, I wound my transformer for the H-Lv2b3 using enameled wire, 0.4 mm diameter on the 50 ohm side and 0.5 mm diameter on the LDMOSs side IIRC, rounding the ferrite holes edges with a drill bit and some sandpaper before and it worked well.

73 de Claudio, DK1CG / IN3OTD

[Bob A. Booey]

Steve,

Speaking of the LPF boards, are the LPF board schematics, layout, etc. on Github?  I was looking for the schematic to see about interfacing the LPF to the Radioberry.

Thanks and 73,

Robert, WA2T

Sent from my iPhone

--

[Nigel Kirlew]

Bob,

You can find it here http://james.ahlstrom.name/hl2filter/

Nigel

N4TKC

To unsubscribe from this group and stop receiving emails from it, send an email to herme...@googlegroups.com.

[rentwist]

Thanks Nigel!  I had run across Jim's page in the recent past but had forgotten about it.

[Steve Haynal]

Hi Robert,

The N2ADR board with build8 has some minor changes versus the one on Jim's site. All files for your N2ADR board are on github. All files for everything HL2 related is on github...

https://github.com/softerhardware/Hermes-Lite2/tree/master/hardware/companions/releases/n2adr_e5b8

73,

Steve

kf7o

[Steve Haynal]

Hi Group,

You should expect 5W or more into a dummy load after the N2ADR filters for all bands below 15M. For 15,12 and 10M, 12 is often the lowest and in the upper 4W range. Sometimes 15 and 10M may dip below 5W. Some other post N2ADR filter power measurements are at:

https://github.com/softerhardware/Hermes-Lite2/wiki/N2ADR-Filter-Board-Plots

Even with the long leads, I am seeing good power output below 15M. I do see less power for 15M and above, similar to what Claudio reported. I did some experiments where I change up various HL2 and filter boards. All measurements are on 12M.

1. Beta5 board with early N2ADR filter board: 4.8W post filter, 5.6W pre filter.
2. Beta5 board with new N2ADR filter board: 4.5W post filter, 5.8W pre filter.
3. Build8 board with long leads and early N2ADR filter board: 3.7W post filter, 4.4W pre filter.
4. Build8 board with long leads and new N2ADR filter board: 3.6W post filter, 4.6W pre filter.
5. Build8 board with rewound T3, new wire and early N2ADR filter board: 4.6W post filter, 5.5W pre filter.
6. Build8 board with rewound T3, new wire and new N2ADR filter board: 4.3W post filter, 5.8W pre filter.

The pre filter load seems to vary, and I don't think it is always 50Ohm. We will have to look at it with the VNA.

I tried switching some of the different brand capacitors but saw no difference.

Rewinding the core with different wire and keeping the leads short made the biggest difference. I used this 30AWG teflon coated silver plated copper wire, an option in the assembly BOM:

https://www.ebay.com/itm/PFEP-Teflon-Silver-Plated-Copper-Wire-Cable-14-16-18-19-20-22-24-26-28-30AWG/263758650100?var=562980843203&hash=item3d693d9af4:g:h8EAAOSwrptbIzV6

We have also used these two types of wire, available from Mouser, MFG P/N 55A0111-24-9  or 650-400R0111-24-9.

We have had problems with enamel coated wire. The insulation fails due to heat and/or edges of the core, and the core conducts.

Even after rewinding, power diminishes with the new N2ADR board versus the old. Power is also slightly lower with the new HL2. The reason may be the ENIG finish. Although this is easier to work with and lead free by simple inspection, it does have higher RF loss. I'm doubtful it would/should matter for HF.  https://www.rogerscorp.com/documents/3313/acm/articles/Impact-of-Final-Plated-Finish-on-PCB-Loss.pdf

In the past, we lowered the preamp output by increasing R55 from 75 to 120 Ohms. We can experiment with reducing this to 100 Ohms to increase overall power output.

I would like to have another week or two of collecting data, but I will probably offer to mail in a flat mailer any interested group participant some of the teflon coated wire and a new resistor for R55. If you go back and look at the original group buy offer, it did say that this is an experimental low margin project and you may need to rewind a core or change a resistor. :)

73,

Steve

kf7o  

[John Greusel]

I was wondering if (lower power out) is not the wire or different capacitors but rather slightly different toroid mix?

On the v8 I received, after fussing with network configurations for a few days, I've come to the conclusion gateware isn't loaded on my unit. I'm working on that now (already got the BSOD in Window 10 with the Chinese USB Blaster :)). I may have it running this weekend.

John

KC9OJV

[James Ahlstrom]

Hello Group,

I doubt the wire could make any difference in power output. But the binocular core could. All my work was done with the larger 202 core, not the smaller core I see with the shipped units. Or the ferrite mix could be off.

The filter board was designed for a power output (after filters) of 5 watts max. If you push that, you will generate more harmonics from the finals, and possibly exceed the legal harmonic output. Steve is right that reducing R55 will easily produce more output. But then you must use the power control settings to reduce the power back to 5 watts for each band.

And before obsessing about power output, please remember that 2.5 watts out is - 3dB, or one half S-unit.

Jim

N2ADR

[ROBERT ENTWISTLE]

Hi Steve,

Thank you for the clarification, "companions" is the key word that I was missing I guess. However, I still cannot find the schematic. The page you linked seems to indicate hermeslite.pdf (which is in another folder) is the schematic but that file covers the HL alone from what I am seeing, doesn't include the LPF schematic?

73,

Robert, WA2T

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/457d2bbf-feaa-436e-8761-307509628b9c%40googlegroups.com.

[in3otd]

Hello,

I'd guess that's a copy/paste error in the README.md file; the board schematic is in the n2adr.pdf file in that folder.

73 de Claudio, DK1CG / IN3OTD

[Steve Haynal]

Hi All,

Sorry about the link typo on the n2adr page. Will try to fix it this weekend. Other can edit the wiki pages with just a free github password.

The core used is the same one we've used for the past couple of builds. It is marked as do not substitute on the BOM, and I don't think Makerfabs made any substitutions for such parts. They ordered directly from Digi-Key or Mouser. Maybe there is some variation from the manufacturer and we are seeing a slightly different mix. It would be interesting if someone wants to do the experiment of swapping the core.

Regarding the wire, what I use is 30AWG with very thin insulation, so the windings are very tight and close together. I'm thinking maybe that makes a slight difference, but it could all be in the noise. Shortening the leads definitely helps the higher frequencies. I agree with Jim that in terms of dB, there is not much difference. Some people really want to see at least 5W on all bands, so I will help them.

You can tell immediately if there is a flashing problem with your board by powering it up with no ethernet connected. All LEDs with be off if there is a gateware problem. Two LEDs will be on/flashing if not. Let me know if you want to make some arrangement with me to have your board flashed. This could be exchanging your board, sending you a loaner working USB blaster, finding someone nearby who is willing to flash your board, etc.

73,

Steve

kf7o

[in3otd]

Hello,

I've been doing some more experiments, still with T3 with shortened leads (but using the same wire): it seems that increasing the output transformer compensation capacitance (C84) helps. As Steve wrote, it could be that the wire used, which has a relatively large insulation, prevents having a good coupling between the windings so that the transformer has a larger leakage inductance. On my H-Lv2b3, I also had my T3 transformer wound with tight windings and the output power at the HF upper end was good.

Here is a graph of my H-Lv2b8 PA output power, without the filter board, for several additional capacitance values added in parallel with C84, with also a graph of a previous measurement on a H-Lv2b3:

with an additional 150 pF I get 5.5 W at 30 MHz (without output filters), adding more capacitance increases the output power until 28 MHz then the power decreases rapidly.
I'll later try to rewind the transformer with a wire with a thinner insulation to see if that brings any further improvements.

[Marty Wittrock]

This is very similar to the power levels I'm seeing on my HL2 as well - thanks for the measurements -

73 de Marty, KN0CK

[John Williams]

Measuring my B8 I found that I had a bad cable. I was using one of those light brown sma jumper cables that are prevalent on ebay. Into my dummy load, the SWR was all over the map. I grabbed a sma jumper that I bought with my 2M HT and the results are much better! This was using spark set at 255 drive, N2ADR hpf-lpf filter setting reading from the spark power calculation.

160M - 5.7W

80M - 5.6W

60M - 5.6W

40M - 4.7W

30M - 5.7W

20M - 5.9W

17M - 4.5W

15M - 4.1W

12M - 2.8W

10M - 4.1W

I am guessing that if I re-wind the PA transformer, the higher bands will improve. I almost never operate there so not an issue I am likely to address.

John W9JSW

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/ec09fbb6-19ce-491f-adf1-35e3cbe21b9a%40googlegroups.com.

[rentwist]

Okay, here's my (CW) power levels using PowerSDR and drive level set to 100%.  I have rewound the PA XFMR (and did not try with original XFMR).  Super short, minimal lead lengths from XFMR to PCB.  The results are kind of surprising based on what others are seeing.  For the power measurements, I'm using a LP-100A just cal'ed by Telepost.  Measurements are into a dummy load.

160m 4.96W

80m 4.78W

60m 4.43W

40m 4.31W

30m 4.60W

20m 5.59W

17m 5.40W

15m 5.46W

12m 5.92W

10m 5.92W

Could my XFMR rewind alone account for this or am I doing something wrong?  PowerSDR is setup per the YT video (and written instructions before that which left out the Apollo setting!).

73,

Robert, WA2T

[rentwist]

The previous measurements were using the TUN button in PowerSDR.  Using the CW key (dot paddle, keyer is not yet working/looked at), I get a bit higher numbers still:

160m 5.09W

80m 4.96W

60m 4.66W

40m 4.49W

30m 4.78W

20m 5.66W

17m 5.53W

15m 5.59W

12m 5.99W

10m 6.05W

While more power is better and I'm happy about that (but why the slight slump at 60/40/30m?), why would the numbers be higher using the CW key (plugged into the HL2) vs. the PSDR TUN button?

73,

Robert, WA2T

[rentwist]

Hold the phone, I only went so far through the HL2 PSDR setup video, so I did not have the filter control set up.  New numbers (TUN/CW now TUN is higher for some reason).  TUN numbers look the same as CW did and CW is a little lower.  Could it be that TUN does not use filters and CW does (now that they are setup to work?

160m 4.96/4.26W

80m 4.72/3.93W

60m 4.43/3.77W

40m 4.26/3.51W

30m 4.55/4.15W

20m 5.594.96W

17m 5.40/4.96W

15m 5.46/5.03W

12m 5.92/4.26W

10m 5.85/5.09W

[Steve Haynal]

Hi Robert,

Did you adjust the maximum PA gain per band as seen at approximate time 5:05 in the video? Also, it would be helpful to try Quisk and SparkSDR too, to make sure there is nothing set incorrectly with PowerSDR. They both run under Windows and Linux.

73,

Steve

kf7o

[Alan Hopper]

Hi Claudio,

I remember the capacitor that was added under the transformer on the vpa line made a difference to power output for me at some frequencies on 3b3, I don't believe the value was ever optimized and  I think stayed at the first value I tried. I wonder if variations in manufacture could have changed whatever happens along the trace from the vpa regulator. Just speculation from a non rf engineer :)

73 Alan M0NNB

[Steve Haynal]

Hi Group,

I experimented with Claudio's C84 increase by 150pF and saw very good results. On two boards where I had rewound T3 with my preferred wire, the power went from about 4.5W to 5W, or a bit less than 0.5 dB, on 12M. Note that this is a 150pF 0805 capacitor placed *on top of* the existing C84, essentially two capacitors in parallel to increase the capacitance. Although this is a small increase in practice, it is inline with past units where power was closer to 5W on all bands.

I also experimented with replacing R55 with a 100 Ohm resistor to increase the RF TX driver output. This added about another 0.5 dB also. I looked at harmonics and they were no worse than before. I didn't look at signal quality (IMD) changes. This mod may be of interest to some, but is not necessary to reach 5W.

For the low RF output on higher frequencies, my recommendation is to rewind T3, with windings as close and tight as possible, and leads as short as possible. Also, add 150 pF in parallel to C84. 

I will send a length of the preferred wire, 150 pF capacitor and 100 Ohm resistor to anyone who bought a build8 unit and would like to make any of these mods. I will use funds from the repair pool. To keep costs low, I will send this in a flat mailer at letter postage rates. If you would like this, please send your Makerfabs order confirmation to hermeslite @ gmail.com. Also, include your preferred address in no more than 5 lines. I will copy and paste this address. If it does not fit, I will request that you fix the address. 

73,

Steve

kf7o

[Gerard Sexton]

---------- Forwarded message ---------
From: Gerard Sexton <seco...@gmail.com>
Date: Sun, 2 Jun. 2019, 13:58
Subject: Re: HL2 Build 8 Observations
To: Steve Haynal <softerh...@gmail.com>

Hi Steve.

Is there likely to be any difference in the output if I use a BN43-202 for T3 (and also using the preferred wire which I already have on order). I have a few spare BN43-202 toroids that I can play with. 

Regards 

Gerard (vk3cg)

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/94ca59d5-5e2f-4ec5-8ce0-d03a3a34a134%40googlegroups.com.

[Steve Haynal]

Hi Gerard,

A BN43-202 core works fine. See the schematic for instructions on using the BN43-202. We used those initially. We switched to the new core for two reasons: lower profile, and availability from Mouser and DigiKey. The new, smaller core actually conducts which requires some caution.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[Sid Boyce]

Hi Steve,
I already have the wire and BN43-202 cores left over from the Beta 5.
Not sure about the 150pF and 100 Ohm right now so I'll let you know
later if I need them.
73 ... Sid.

On 02/06/2019 05:33, Steve Haynal wrote:
> Hi Gerard,
>
> A BN43-202 core works fine. See the schematic for instructions on
> using the BN43-202. We used those initially. We switched to the new
> core for two reasons: lower profile, and availability from Mouser and
> DigiKey. The new, smaller core actually conducts which requires some
> caution.
>
> 73,
>
> Steve
> kf7o
>
>
> On Saturday, June 1, 2019 at 9:02:19 PM UTC-7, Gerard Sexton wrote:
>
>
> ---------- Forwarded message ---------
> From: *Gerard Sexton* <seco...@gmail.com <mailto:seco...@gmail.com>>
> Date: Sun, 2 Jun. 2019, 13:58
> Subject: Re: HL2 Build 8 Observations
> To: Steve Haynal <softerh...@gmail.com
> <mailto:softerh...@gmail.com>>
>
>
> Hi Steve.
> Is there likely to be any difference in the output if I use a
> BN43-202 for T3 (and also using the preferred wire which I already
> have on order). I have a few spare BN43-202 toroids that I can

> play with.Â
> RegardsÂ

> Gerard (vk3cg)
>
> On Sun, 2 Jun. 2019, 13:36 Steve Haynal, <softerh...@gmail.com

> <mailto:softerh...@gmail.com>> wrote:
>
> Hi Group,
>
> I experimented with Claudio's C84 increase by 150pF and saw
> very good results. On two boards where I had rewound T3 with
> my preferred wire, the power went from about 4.5W to 5W, or a
> bit less than 0.5 dB, on 12M. Note that this is a 150pF 0805
> capacitor placed *on top of* the existing C84, essentially two
> capacitors in parallel to increase the capacitance. Although
> this is a small increase in practice, it is inline with past
> units where power was closer to 5W on all bands.
>
> I also experimented with replacing R55 with a 100 Ohm resistor
> to increase the RF TX driver output. This added about another
> 0.5 dB also. I looked at harmonics and they were no worse than
> before. I didn't look at signal quality (IMD) changes. This
> mod may be of interest to some, but is not necessary to reach 5W.
>
> For the low RF output on higher frequencies, my recommendation
> is to rewind T3, with windings as close and tight as possible,
> and leads as short as possible. Also, add 150 pF in parallel

> to C84.Â

>
> I will send a length of the preferred wire, 150 pF capacitor
> and 100 Ohm resistor to anyone who bought a build8 unit and
> would like to make any of these mods. I will use funds from
> the repair pool. To keep costs low, I will send this in a flat
> mailer at letter postage rates. If you would like this, please
> send your Makerfabs order confirmation to hermeslite @

> gmail.com <http://gmail.com>. Also, include your preferred

> address in no more than 5 lines. I will copy and paste this
> address. If it does not fit, I will request that you fix the

> address.Â
>
> 73,
>
> Steve
> kf7o
>
>
> Â

>
> On Saturday, June 1, 2019 at 1:47:44 PM UTC-7, Alan Hopper wrote:
>
> Hi Claudio,
> I remember the capacitor that was added under the
> transformer on the vpa line made a difference to power
> output for me at some frequencies on 3b3, I don't believe

> the value was ever optimized and  I think stayed at the

> first value I tried. I wonder if variations in manufacture
> could have changed whatever happens along the trace from
> the vpa regulator. Just speculation from a non rf engineer :)
> 73 Alan M0NNB
>
> On Friday, May 31, 2019 at 2:36:44 PM UTC+1, in3otd wrote:
>
> Hello,
> I'd guess that's a copy/paste error in the README.md
> file; the board schematic is in the n2adr.pdf

> <https://github.com/softerhardware/Hermes-Lite2/blob/master/hardware/companions/releases/n2adr_e5b8/n2adr.pdf>

> it, send an email to hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit
> https://groups.google.com/d/msgid/hermes-lite/94ca59d5-5e2f-4ec5-8ce0-d03a3a34a134%40googlegroups.com

> <https://groups.google.com/d/msgid/hermes-lite/94ca59d5-5e2f-4ec5-8ce0-d03a3a34a134%40googlegroups.com?utm_medium=email&utm_source=footer>.

> For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hermes-lite/c60beb3f-4081-4a22-81eb-92554b05036b%40googlegroups.com
> <https://groups.google.com/d/msgid/hermes-lite/c60beb3f-4081-4a22-81eb-92554b05036b%40googlegroups.com?utm_medium=email&utm_source=footer>.

> For more options, visit https://groups.google.com/d/optout.

--
Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot
Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support
Senior Staff Specialist, Cricket Coach
Microsoft Windows Free Zone - Linux used for all Computing Tasks

[rentwist]

Hi Steve and all,

Okay, I completed the settings in the video all the way through.  Here are my latest TUN/CW power readings on the LP-100A:

160m 6.12/5,34W

80m 6.54/5.59W

60m 6.19/3/5.66W

40m 5.99/5.03W

30m 6.12/6.19W

20m 6.12/5.99W

17m 5.79/5.53W

15m 5.46/5.03W

12m 5.92/4.26W

10m 5.85/5.09W

These numbers look even better than what I was seeing before I completed the setup properly, only 12m using CW key looks a tad low, the others look surprisingly high given what other are seeing.  Only change is rewound PA XFMR with 0.5mm enameled wire secondary, 22 AWG Teflon wire primaries and super short lengths.  I do plan on redoing the XFMR again using Steve's recommendations.

Do not understand why the TUN values differ from values using a CW key.  Speaking of the CW key, I cannot seem to get the keyer function to work (given the settings I see, there is a keyer function, right?)  Left paddle just straight keys and right paddle keys but no power out.  I have looked high and low in the PSDR settings.  Anyone have advice on this?

I seem to recall that the N2ADR LPF board SWR bridge function auto selects the higher of the two power values for forward, something like that?  So, when I saw no power output indicated in PSDR (even though the HL2 is putting out power) in Tx, I selected Ref Pwr under TX Meter in Tune section.  Now I get ~100W in the meter section when in TX.  I played with the PA Settings > Wattmeter table for the 100W Meter Trim but it does not seem to have an effect.  In short  not having much luck getting the TX meter to read actual Tx power, I think this should be possible.

73,

Robert, WA2T

[rentwist]

Regarding CW, maybe I am not "doing it right".  Thinking maybe the 3.5mm jack maybe is for a mic (and not a CW key)?  Or does the jack function as both a Mic and CW key jack?  If no, how does one go about hooking up a CW paddle?

[John Williams]

Definitely not a mic...

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/d020b85a-a8e5-46c0-b80c-b58785ed22f7%40googlegroups.com.

[John Williams]

In PowerSDR the fwd-ref are swapped and the magnitude is about 11X hotter. I build my power sensing board to exact specs from an Alex design and it ran perfect in PowerSDR, but was 11X low in Spark. Never tried it in Quisk. The pick the highest value as forward algorithm caused me to think the polarities were ok when I tested my board with Spark.

See my QRZ page for a pic. I had Jim add an external connector so I can still use it. Will hook it up soon to see how it works.

John

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/6f109e4e-da35-49de-81a0-92b29c3e3fec%40googlegroups.com.

[in3otd]

Hello,
as far as I understand the connections to the CW key jack (CN4) are
TIP : CW (straight) key input
RING : PTT input

73 de Claudio, DK1CG / IN3OTD

[dick_...@hotmail.com]

The phenomena you are experiencing is a direct function of a directional coupler [note the word directional].  There is no real "standard" as to which direction is "normal'. I have viewed your QRZ page [nice layout] and the solution is simple in theory, but may not be practical with multiple pieces of software. 

Simply reverse the input and output !

Do you have a link to the design files for the bridge that you developed ?

For further explanations refer to my web page

http://www.k9ivb.net/Directional_Coupler/

The Michael Ellis link at the top of the page will take you to his complete technical details.

or just download the Schematic which has most of the information with references

http://www.k9ivb.net/Directional_Coupler/files/Directional%20Coupler.pdf

Dick K9IVB

To unsubscribe from this group and stop receiving emails from it, send an email to herme...@googlegroups.com.

[Steve Haynal]

Hi All,

Responding to several posts in this thread. 

First, I don't think there is anything inferior about the core that is included with the HL2 which would make it better to replace with a BN43-202. I am just saying that a BN43-202 will also work.

Second, Claudio is correct about CW. The thinking is that most people have a keyer they like so can just use the tip of the connector (CW on/off). Others wanted an external PTT, hence the second connection. My philosophy has been that this is a "back room" radio that is not right in front of the operator but off in an equipment room. Hence there is limited LEDs and connectivity. This also keeps the price down. If someone is looking for a project, it should be possible add back the openHPSDR keyer and repurpose the PTT on the connector.

Third, I think Jim has an option to just reverse the windings of the power pickup toroid. You can remove the toroid, keep the windings in place, but just rotate it by 180 degrees. This should then switch FWD/REV??? Jim can speak better to this. I'm also willing to swap this in the gateware if the default from Makerfabs appears to swap FWD/REV for PowerSDR.

73,

Steve

kf7o 

[Steve Haynal]

Hi Robert,

Be careful with enamel wire and the T3 core. The core conducts and we've had cases in the past where over time and exposure to temperature, the enamel breaks down and there is a short.

73,

Steve

kf7o

[in3otd]

Hello,
I've rewound T3 on my H-Lv2b8 like the transformer on my H-Lv2b3 and now the PA output power is virtually the same between the two boards:



so the core included in the build 8 works as expected. The measurement above was done without additional capacitors at C84.
I've also characterized the transformer before and after rewinding it and can confirm that the original transformer has a slightly higher leakage, which causes a droop in the upper HF frequency response.
I wound the transformer using a 0.4 mm enameled wire since this is what I used for the beta 3 measurements I have as reference. I've wound a small strip of kapton tape around the transformer center leg as additional insulation to avoid any damage to the enameled wire.
I did not readily find any small Teflon-insulated wire in my junkbox to check if a wire with thicker insulation could have a slightly worse response but I'll check again and wind/test a new transformer if I find a suitable wire.

73 de Claudio, IN3OTD / DK1CG

[Bob A. Booey]

Roger that Steve, I wrote you asking for the wire, etc. to be included with my replacement LPF board, I will rewind the transformer again with that wire.  Thanks.

Sent from my iPhone

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/766f353c-c838-438e-b691-3133ace3607d%40googlegroups.com.

[Bob A, Booey]

Interesting on the CW, I guess the keyer stuff in PSDR led me (incorrectly) to think there was a keyer present in the HL2.  So, following the logic of a back room radio, one would still need to be able to connect to the CW key line in some fashion, I'm guessing there is no facility or easy way to connect a keyer to the host PC then?  Us 99% CW guys will want to know what options we have in this regard.  Along the same lines, I guess audio for voice/digital modes comes from the host PC then?  Is there a way to connect a hand mic to a host PC?  Guess the 3.5mm mic jack would do for mic audio, but probably not for PTT?  I know (now) that there are two PTT lines (I think the RCA jack on the LPF board is PTT also?) on the HL2 box.

Personally, I would love to have the keyer functionality added into the HL2 but I am probably not the guy to do that.  I feel confident when I say that I would need a lot of hand holding to get the job done!

Thanks and 73,

Robert

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/2878719b-dabf-411f-b15a-d56b1cc89c14%40googlegroups.com.

[John Williams]

Hi Dick,

Actually you developed the bridge :-)

You will recognize the circuit as it came from here - http://www.k9ivb.net/Munin2/Alternate_Alex_Interface_Ver6.2.pdf

My board information is on github here called FR Sense Board - https://github.com/W9JSW/Hermes-Lite/tree/Amplifiers

I figured that it was easier to use this design than one of my own and then have to re-calculate the values to get it to jibe with Hermes/Alex.

I have the board gerbers uploaded to oshpark. There is a minor problem with them in that the silk screen for the diodes is reversed. Otherwise they build and run FB. Never did a China run and I bet they would be dirt cheap there.

https://oshpark.com/shared_projects/z3pei6Ab

John

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/db0db8f9-4ddf-439d-864a-10fe32cb5488%40googlegroups.com.

[Steve Haynal]

Hi Claudio,

Thanks for the analysis. All my measurements have been with a N2ADR board in place. I've noticed about 0.2 to 0.4 dB more loss in the newest N2ADR boards. Did you purchase a N2ADR board, and if so, do you see any differences? Still, the output is around 5W on all bands with rewinding the core and adding the extra 150pF. 

If you add the 150pF to your current setup, does it improve high frequency further? Do you see any downside to adding the 150pF?

73,

Steve

kf7o 

[Steve Haynal]

Hi Robert,

The PTT on the back is an output. It is used to send PTT to an external PA. The HL2 pulls the line to ground when PTT is activated.

The PTT on the front is an input. It is an alternate way to trigger PTT that some people wanted.

You can do CW through the host, but the latency is much longer, no QSK.

All audio (mic and audio output) is through the host PC for the HL2. This is different from openHPSDR radios.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[rentwist]

Thank you Steve.

Okay, so I want CW direct to the HL2 then.  I'll press a keyer into service for now.  Maybe I can get myself oriented toward working on the openHPSDR keyer incorporation you mentioned, that would be quite the learning experience for me but I think I could handle it.  Question is if every one I would bother trying to pull myself up by my bootstraps could handle it! ;-)

73,

Robert, WA2T

[James Ahlstrom]

Hello Group,

Regarding the forward/reverse power sensor, the polarity depends on the winding sense.  That is, if you wind the toroid one way (call it left handed) the forward power is on one side (call it A). But if you wind the toroid left-handed, the forward power is on B. Since the toroid is hand wound, there is no standard winding direction.

So the solution in Quisk is to simply assume the higher of the two power readings is "forward" and compute accordingly. There is no point in reversing any windings.

Jim

N2ADR

[James Ahlstrom]

Hello Group,

For those experiencing abnormally low power output, you could test to see if the filter board is the problem. Just change all the filter bits to zero. Then the RF power is routed straight through without any filters. The indicated power will be incorrect, since it measures all harmonics too, but it is worth a try.

Another filter board test is to run the HL2 at maximum power for a while, stop Tx, and test the filter board filters with your finger (carefully) to look for heat that may indicate a problem. But if the problem is incorrect filter component values, the filter may not heat even though the reactance will cause an in-band loss due to mismatch.

Jim

N2ADR

[in3otd]

Hello,
I did purchase also a N2ADR filter board, I've done some measurements on that also but only with a VNA (so at low power) for the moment. The board does seem to have slightly higher losses than the E1 prototype I have but not much.
I've tried to make an animated gif image showing the filters responses for the two boards:



the Makerfabs board seems to have lines with a slightly higher impedance, from the measurements, but this alone does not explain the losses differences. I know that the E1 was done with a slightly different/better substrate (FR402) but where there some changes to the main RF traces layout between the two boards?

The overall C84 optimal value depends on the transformer leakage, a higher the leakage will need more capacitance to flatten the response but when the leakage is too high it will not be possible to have a flat(ter) response extending up to the higher frequencies.

73 de Claudio, DK1CG / IN3OTD

[in3otd]

...nope, animated gifs seems a too old technology for Google Groups, so here are two separate graphs with the filters responses:

N2ADR E1 prototype:

N2ADR E5b8 from Makerfabs:

[Steve Haynal]

Hi Jim,

The recommendation of reversing the windings was to accomodate PowerSDR, which does not sense and switch to the higher output like Quisk does.

73,

Steve

kf7o

[Steve Haynal]

Hi Claudio,

Thanks for the measurements. The new board appears to be ~0.2dB lower on the higher frequencies. This would appear to be a board or relay issue as the pass thru also shows the drop. The biggest board change was ENIG finish, but I think you are comparing against an OshPark prototype which also had ENIG finish. There were a couple of other minor changes, but nothing that I suspect. The losses are very small, so perhaps just due to board variations. I don't think it is large enough to worry about, especially since people can increase power output in other ways.

73,

Steve

kf7o

[Steve Haynal]

Hi Group,

The requested mod kits should go out in the mail tomorrow. The 150pF capacitor is the white unmarked device, and the 100 Ohm resistor is black and marked. Please remember that the capacitor goes "piggy back" on top of C84. There is 30cm of wire. I measured 10.5cm required for the 4 windings and 3.7cm each for the two single windings, assuming 2.5mm is stripped. 

I will send out more mod kits next week if there are any more requests.

73,

Steve

kf7o

  

On Saturday, June 1, 2019 at 8:36:09 PM UTC-7, Steve Haynal wrote:

Hi Group,

I experimented with Claudio's C84 increase by 150pF and saw very good results. On two boards where I had rewound T3 with my preferred wire, the power went from about 4.5W to 5W, or a bit less than 0.5 dB, on 12M. Note that this is a 150pF 0805 capacitor placed *on top of* the existing C84, essentially two capacitors in parallel to increase the capacitance. Although this is a small increase in practice, it is inline with past units where power was closer to 5W on all bands.

I also experimented with replacing R55 with a 100 Ohm resistor to increase the RF TX driver output. This added about another 0.5 dB also. I looked at harmonics and they were no worse than before. I didn't look at signal quality (IMD) changes. This mod may be of interest to some, but is not necessary to reach 5W.

For the low RF output on higher frequencies, my recommendation is to rewind T3, with windings as close and tight as possible, and leads as short as possible. Also, add 150 pF in parallel to C84. 

I will send a length of the preferred wire, 150 pF capacitor and 100 Ohm resistor to anyone who bought a build8 unit and would like to make any of these mods. I will use funds from the repair pool. To keep costs low, I will send this in a flat mailer at letter postage rates. If you would like this, please send your Makerfabs order confirmation to hermeslite @ gmail.com. Also, include your preferred address in no more than 5 lines. I will copy and paste this address. If it does not fit, I will request that you fix the address. 

73,

Steve

kf7o

 

On Saturday, June 1, 2019 at 1:47:44 PM UTC-7, Alan Hopper wrote:

Hi Claudio,

I remember the capacitor that was added under the transformer on the vpa line made a difference to power output for me at some frequencies on 3b3, I don't believe the value was ever optimized and  I think stayed at the first value I tried. I wonder if variations in manufacture could have changed whatever happens along the trace from the vpa regulator. Just speculation from a non rf engineer :)

73 Alan M0NNB

On Friday, May 31, 2019 at 2:36:44 PM UTC+1, in3otd wrote:

Hello,

I'd guess that's a copy/paste error in the README.md file; the board schematic is in the n2adr.pdf file in that folder.

73 de Claudio, DK1CG / IN3OTD

On Friday, May 31, 2019 at 3:04:24 PM UTC+2, ROBERT ENTWISTLE wrote:

Hi Steve,

Thank you for the clarification, "companions" is the key word that I was missing I guess. However, I still cannot find the schematic. The page you linked seems to indicate hermeslite.pdf (which is in another folder) is the schematic but that file covers the HL alone from what I am seeing, doesn't include the LPF schematic?

73,

Robert, WA2T

On May 31, 2019 at 1:26 AM Steve Haynal <softerh...@gmail.com> wrote:

Hi Robert,

The N2ADR board with build8 has some minor changes versus the one on Jim's site. All files for your N2ADR board are on github. All files for everything HL2 related is on github...

https://github.com/softerhardware/Hermes-Lite2/tree/master/hardware/companions/releases/n2adr_e5b8

73,

Steve

kf7o

[RHQQ2YXRKT]

 

Hi Group,

 

I am getting ready to try transmit.

 

1.  Clicking /radio settings/ filters/      Filter Type is set to J16.   I have the N2ADR filter board,  I am guessing that needs to be N2ADR-hpf-lpf  ?

 

2.  I have not used the area under the two PA transistors to heat sink to the case.  I have a copper bar touching the tops of the PA transistors directly which is bolted to the case. 

     I am using the recommended case with the PCB in the slide in groove.

 

For best RF stability what is the best part of the PCB track to make a connection with the aluminium case ?

 

I would be interested in any views.

 

John G3UGY

 

 

 

 

 

 

[rentwist]

What SDR software are you using?

Interestingly, I believe PiHPSDR refers to J16 also in the OC menu (going by memory).  Do not know how (or if) that can be changed in the case of PiHPSDR.  I thought J16 was a reference to a (virtual?) connector, perhaps not.

73,

Robert, WA2T

[John Williams]

J16 is the gateware feature that controls the filters in hermes lite. In hl1 the lines were directly available at the frontend. For hl2 Steve chose to map that onto an i2c bus to go off board. So all software provides a j16 facility to control any type of filters you want to design. N2ADR is one such design. Alan and James make it easy to fill in the bits for J16 as the N2ADR card is now ubiquitous. 

John W9JSW 

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/82bef516-599f-4780-bc92-dc8beb58cffc%40googlegroups.com.

[ROBERT ENTWISTLE]

Excellent to learn, thanks John!

73,

Robert, WA2T

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CADyLd7x4vh4uBXD%3DdfhbvRPk5Aw3CmejBbR7U_Mn%3D0fdOSSNPQ%40mail.gmail.com.

[Alan Hopper]

John,

I guess this is in Spark, so yes any of the N2ADR options are good, they all filter correctly on tx and the options just affect receive, I need to improve the doc on this.

73 Alan M0NNB

[RHQQ2YXRKT]

 

Alan,

 

For the filters to affect tx and rx which option shall I select?

 

J16   /  N2ADR-hpf-lpf  /  N2ADR-lpf  /  N2AR-no-rx-filter

 

 

73 .... John

 

 

 

From: 'Alan Hopper' via Hermes-Lite

Sent: Wednesday, June 5, 2019 7:50 PM

To: Hermes-Lite

Subject: Re: HL2 Build 8 Observations

[Alan Hopper]

John,

go for any of the N2ADR ones, N2ADR-hpf-lpf should probably be the default as it filters as much as it can on receive, this may or may not be optimum if you live somewhere quiet. The default J16 settings have no filtering on tx or rx.

73 Alan M0NNB 

[Duncan Clark]

In message <d2d5ec87-ee48-42e5...@googlegroups.com>,
Steve Haynal writes

>Please remember that the capacitor goes "piggy back" on top of C84.

Where is C84? Darned if I can see it from the various board photos.

Mine board is in it's enclosure so Murphy's law says bottom of the
pcb :-)

I haven't yet worked out how I can view a pcb overlay from the github.
Is there a pdf of the overlay anywhere?

Thanks

Duncan
--
Duncan Clark
G4ELJ

[Gerard Sexton]

It is on the bottom...

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/LjJJneCytD%2BcFwpW%40ntlworld.com.

[in3otd]

Hello,
there's an interactive BOM/layout viewer for the build7 at http://www.hermeslite.com/hl2b7-ibom.html - it should be practically identical to the build 8.
The build 8 file is at https://github.com/softerhardware/Hermes-Lite2/blob/master/hardware/hl/releases/hl2p0build8/ibom.html but it seems you have to save it locally to make it work (Steve, can you upload this too to the website?).
Anyway, C84 is on the bottom, between the T3 primary winding holes.

73 de Claudio, DK1CG / IN3OTD

On Thursday, June 6, 2019 at 12:00:41 AM UTC+2, Duncan Clark wrote:

In message <d2d5ec87-ee48-42e5-9556-85dc90...@googlegroups.com>,

[Steve Haynal]

Hi,

The bottom edge of the PCB just by the two LDMOS devices in the PA is the most important place to have good thermal contact with the slot of the enclosure. It helps to use a little thermal paste here, and provide some contact pressure with a nut and bolt.

73,

Steve

kf7o

[RHQQ2YXRKT]

 

Hi Steve,

 

I have been trying TX.   The PCB’s are in the recommended case. I have a 50 Ohm dummy load and Watt meter connected.

 

Mode AM (so as to produce a carrier) band 14MHz,  Drive level slider to minimum.

 

Pressing the PTT in Spark PA current shown is 175mA  Forward and reflected power are zero (current draw from the 12V linear power supply is 400mA)

 

Increasing the drive slider even to maximum doesn’t increase the pa current from 175mA.  Power supply current draw stays at 400mA.

 

I want to try running with the PCB’s out of the case but that means there is no heat sink. 

How long is it safe to run the PA without the heat sink?

 

Can you give a link to the PCB component layout?

 

John G3UGY

[Alan Hopper]

Hi John,

have you enabled the pa in radio settings / pa control? it defaults to off.

73 Alan M0NNB

[Steve Haynal]

Hi John,

You should be able to run for short periods of time and/or at low power with the PCB out of the enclosure. It looks like you have are seeing the standing bias current with no modulation for a cold unit. I would try with tune or WSPR/FT8 out on SparkSDR just to make sure you have signal. AM has additional complexity in setting up the audio properly from your mic to the software.

All the information for build8 is in the release directory:

https://github.com/softerhardware/Hermes-Lite2/tree/master/hardware/hl/releases/hl2p0build8

Use gerbv to view the gerber files:

http://gerbv.geda-project.org/

Claudio made a recent post pointing to the interfactive BOM. That is very useful for locating components. Build7 and build8 are identical in terms of component placement.

73,

Steve

kf7o

[RHQQ2YXRKT]

Hi Steve,

 

It looks like there is an intermittent.

 

With the pcb’s out of the case, 14 MHz AM using the Tune button (I am guessing that is the phone icon in Spark) with the drive slider mid position, mostly when you hit tune you just see the PA quiescent current of about 180mA.  Very occasionally PA current shown in Spark goes up to around 500mA approx and Spark shows full amplitude signal peak in the centre of the frequency display. 

 

I unplugged the filter pcb a couple of times but it didn’t make any difference.

 

John G3UGY 

 

    

 

 

 

From: Steve Haynal

Sent: Friday, June 7, 2019 4:52 AM

To: Hermes-Lite

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/c427e0ab-f15d-46e1-8f91-e4cfad8bec39%40googlegroups.com.

[Steve Haynal]

Hi John,

I'm not sure what should happen with what you are doing. Maybe Alan can comment. I would still like to remove the AM. Can you try Quisk? The video covers how tune works with Quisk. Watch the whole video, and around 9:40 there are some spot power measurements.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[John Williams]

Sparks equivalent to spot is sig, down near the bottom of the pulldown.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/c427e0ab-f15d-46e1-8f91-e4cfad8bec39%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

 

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/42d3883f-af06-475f-ba37-1e28ba99efa1%40googlegroups.com.

[Alan Hopper]

John beat me to it,

yep sig is the place to start, the default settings should give you a full power signal.  The telephone is indeed the ptt button.

73 Alan M0NNB 

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/c427e0ab-f15d-46e1-8f91-e4cfad8bec39%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to herme...@googlegroups.com.

[in3otd]

Hello,
I agree that the slightly higher losses of the new filter board are not due to the ENIG finish - besides being used on both versions that finish is also present only on the exposed copper; under the solder maks there is only bare copper, so ENIG is only on a very small part of the traces. Also usually the higher ENIG losses are seen at higher frequencies.
I noticed a difference in the layout between the E1 (actually E3 but I think it's the same) and the E5b8: the ground clearance is larger on this latter, which could explain why the traces impedance is more around 60 ohm rather than 50 ohm.
Again, it may not be the main reason of the slightly higher losses but it would be good to center those random variations around the right mean, hi.

73 de Claudio, DK1CG / IN3OTD

[Alan Hopper]

Hi John, all,

Spark will show this intermittent TX behaviour if you try to transmit a voice mode with either the audio muted or 'none' selected as audio input.

The most common cause of 'none' being selected for the audio input is not having the audio device defaulted to 48KHz in the windows audio settings.

73 Alan M0NNB

ps the next release will clean up the intermittent behaviour.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[RHQQ2YXRKT]

 

Hi Alan,

 

It looks promising, TX’ing appears to be working fine now.  I’ll do some more testing to confirm.

 

One thing I noticed, when receiving in usb or lsb and I switch to NFN or FM the frequency band scope freezes and receive noise stops (no mute symbol) when I click usb or lsb the band scope unfreezes and there is receive noise.

 

John G3UGY.   

 

 

 

From: 'Alan Hopper' via Hermes-Lite

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/8e7b2109-acea-457e-b2a8-d97645d57aaa%40googlegroups.com.

[Alan Hopper]

John,

that's good to hear.

fm and nfm only work when the radio bandwidth is set to 384KHz, It is on my list to at least add a warning message to these modes and maybe relax the requirement for nfm, 

73 Alan M0NNB

[RHQQ2YXRKT]

 

Thanks for your help Alan.

 

There was no problem with hardware or software, just a matter of knowing the setting up procedure.

 

73 John G3UGY

 

 

 

From: Alan Hopper

Sent: Saturday, June 8, 2019 2:17 PM

To: RHQQ2YXRKT

Subject: Re: HL2 Build 8 problem with TX

John,

that's good to hear.

fm and nfm only work when the radio bandwidth is set to 384KHz, It is on my list to at least add a warning message to these modes and maybe relax the requirement for nfm,

It is probably best to keep these communications on the Hermes Lite or SparkSDR google groups so others can benefit, you won't be the last to have these issues:)

73 Alan M0NNB

 

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/c427e0ab-f15d-46e1-8f91-e4cfad8bec39%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Steve Haynal]

Hi Claudio,

Interesting. Maybe this happened when I imported the files into a new version of KiCAD. This is a 4 layer board, so would think the ground plane would be the biggest factor. For the next run, I'll go for the 0.2mm clearance again. I've opened an issue on github: https://github.com/softerhardware/Hermes-Lite2/issues/116

73,

Steve

kf7o

[RHQQ2YXRKT]

 

I have just tested the RF power output on the HL using a Marconi 2503 50 Ohm RF thermocouple DC to 1GHz power meter.

The manual shows that the meter can be calibrated using a known DC voltage.

 

Tests on 1.900 MHz with power slider at maximum.

 

SIG    PA current is 1535mA  RF power ouput  6.5W 

AM    PA Current is  760mA   RF power output   2W 

NFM  PA current is 1412mA   RF power output   5W

 

 

73 John G3UGY

 

 

 

 

[RHQQ2YXRKT]

 

Hi Alan,

Are these figures what you would expect?

 

73  John G3UGY 

 

 

 

 

 

 

From: Alan Hopper

Sent: Saturday, June 8, 2019 2:17 PM

To: RHQQ2YXRKT

Subject: Re: HL2 Build 8 problem with TX

John,

that's good to hear.

fm and nfm only work when the radio bandwidth is set to 384KHz, It is on my list to at least add a warning message to these modes and maybe relax the requirement for nfm,

It is probably best to keep these communications on the Hermes Lite or SparkSDR google groups so others can benefit, you won't be the last to have these issues:)

73 Alan M0NNB

 

On Sat, Jun 8, 2019 at 2:01 PM RHQQ2YXRKT <rhqq2...@googlemail.com> wrote:

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/c427e0ab-f15d-46e1-8f91-e4cfad8bec39%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Alan Hopper]

John,

these sound about right, Sig uses pretty much the full range of the dac,  fm mode is a bit cautious as the final filtering can create clipping if you try to use the full dac range 
 before them, am uses 1/2 the dac range with no modulation so that is slightly higher than I would expect, I think I may have a bug in generating my AM carrier.

73 Alan M0NNB

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/c427e0ab-f15d-46e1-8f91-e4cfad8bec39%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.