Lost in 100D signal path

marc
Posts: 34
Joined: Mon May 15, 2017 2:40 pm

Lost in 100D signal path

Postby marc » Sun Jun 11, 2017 2:52 pm

Hi
I’m trying to understand the incoming (receiving) signal path of the Angelia/100D… and I cannot kill the 3rd level Boss. Where am I making a mistake ?
Using the 100D schematic (100W PA Rev 24)
Signal enters at Ant(1/2/3)==>goes thru the LPF==>gets out of the LPF (label “line LPF_IN”) ==>crosses the TX_RX_Relay (label “RX_MASTER_IN”)==>arrives at the hpf input/and RX BNC Junction (labels “XVTR_IN/RX2_IN/RX3_IN”)==>goes thru Attenuator==> goes thru hpf==>goes thru K19 RX/PS relay (label “RX1LINE”)==>goes thru K36 and J13 (To RX on Angelia board) or to external BNC label “RX_BYPASS_OUT”, depending on the D3 bit (also RX_BYPASS_OUT) fired by the SPI.

If K36 is activated, the whole hpf/lpf is discarded , the Bypass input goes directly to the first ADC (or worth : if I follow the documentation, I should be able to insert an amp or filter that will be “looped” with the HPF section… circular error, abort, ignore, retry ?)

The original Alex schematic (page 24 of 38) and the Anan documentation (Anan 100D/200D manual, page 60 of 71) tell me a totally different story. The "Bypass out" relay is located between the HPF and LPF, it's input comming from the lpf ou, the "normal" output beeing in // with other RX bnc inputs, and the other output going to the rear panel, thus allowing to put an amp or filter inserted between the lpf and the hpf. And this seems pretty much logical would say Mr Spock.

What reading error am I committing ? (can't believe there is an error in the 100D schematic labeling... Pebcak)
73’
Marc f6itu
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Lost in 100D signal path

Postby w-u-2-o » Sun Jun 11, 2017 6:42 pm

Marc,

It's not surprising that you are confused. The schematic for the Rev. 24 board is obviously correct, while the block diagram in the user manual is old and only refers to the older Rev. 15 and 16 boards.

Here are the differences:

On the Rev. 15 and 16 boards the RX output could be sent out the Bypass connector for an external filter and then brought back into either the Ext 1 or 2 inputs on receive. On the Rev. 24 board I believe you now have to use the XVRTR and Ext connections in a different manner, although I've never done it.

On the Rev. 15 and 16 boards one used the Ext 1 or 2 input for external PureSignal feedback. Now one must use Bypass.

On the Rev. 15 and 16 boards there was no explicit internal path for PureSignal feedback, one had to use the existing crosstalk on the board. On the Rev. 24 there is now K19 to provide an explicit and dedicated path.

On the Rev. 15 and 16 boards there was a relay that grounded the RX2 input during TX. That relay no longer exists on the Rev. 24 board, thereby allowing one to monitor their own transmitted signal on RX2 in full duplex. However one must be careful not to overload this input during TX (use an external limiter).

73,

Scott
marc
Posts: 34
Joined: Mon May 15, 2017 2:40 pm

Re: Lost in 100D signal path

Postby marc » Mon Jun 12, 2017 6:42 am

Hi Scott
Thanks again for these very clear explanations. I couldn’t find anything looking for a “changelog” file giving these infos. I’ll modify my personal version of the manual to take into account these mods.
Just one last stupid question (I hope) to complete the “big picture”
Where the XVTR TX signal (rear panel) comes from ? (J3 /Angelia output I imagine) and how, in this case, is switched off the bias of the amp -or any other way to avoid frying the local amp without any load?
Personal thought: if XVTR is software selected, the “TXRX_Relay” line is disabled, thus cutting the bias, the unused TX path and solving the problem.
Thanks again for your patience and help.
73’
Marc f6itu
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Lost in 100D signal path

Postby w-u-2-o » Mon Jun 12, 2017 4:58 pm

Marc,

You are correct about the source of the transverter output.

As for the switching logic when using transverters I'm not entirely sure. I don't employ any transverters here. What you wrote makes sense, of course.

Since this entire thread has nothing to do with firmware but it currently in the firmware sub-forum, I'm going to move it to the transverter sub-forum and perhaps someone who is more familiar with transverter operations can answer your questions.

73,

Scott
marc
Posts: 34
Joined: Mon May 15, 2017 2:40 pm

Re: Lost in 100D signal path

Postby marc » Mon Jun 12, 2017 7:08 pm

Once again, thank you for your answer Scott
I feel a little bit stupid asking you what seems obvious to you (and all other SDR users).
Firmware and hardware in the whole Hermes line looks... as entangled as a pair of Qubits in quantum communications :-D
I may be wrong, but I think that a good documentation index in the "User Guides, Schematics & Other Documentation" section would help a lot

Most of my questions would probably find an answer in
- a general "changelog" hardware file (and associated software)
- an "up to date" SPI protocol description, a list of the different signal labels -and associated functions- used in the schematics, a functional flowchart for each version.

Well… not an easy task, one flowchart or logic diagram for each aspect of the rig : hardware, firmware and client software… it’s an almost impossible mission. A general logic/hardware diagram of the last existing version could be enough for most of us. Any other question would be solved by a “RTFcode”
VY 73’
Marc f6itu
marc
Posts: 34
Joined: Mon May 15, 2017 2:40 pm

Re: Lost in 100D signal path

Postby marc » Sat Jan 20, 2018 2:35 pm

... a little bit less lost but ....

Two relays are still posing problems in my understanding of the 100D architecture : K19 and K36 (Rev 24 board)

As far as I understand, K19 forwards the HPF output to RX1_Line when receiving, and inject the PS feedback signal in the receiver when transmitting. But not directly : RX1_Line is NOT a label designating one of the two ADC input, but just an internal wiring going to....

... K36. Which takes any external signal coming from the external plug "RX Bypass Out" OR from this aformentioned "RX_Line" and forward it to (by default) ADC1 input.

In other words, depending on the parameters given by PowerSDR (particularly the check box "bypass on TX" in the Alex antenna setup screen) , this switching allows to select an "internal" (PS_Feedback from the internal Tandem Match) or "external" pure signal input when transmitting, and the hpf output to the first receiver (physically ADC1) during RX, via K36.

When the checkbox is clear, RX_Bypass_Out is useless (a 50 Ohms load could even be a good idea on this input), K36 is just acting as a bare and straight wire

Am I right ?

73'
Marc f6itu
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Lost in 100D signal path

Postby w-u-2-o » Sat Jan 20, 2018 3:29 pm

That is exactly right, Marc. You are not lost at all.

The signal names are certainly very confusing on the schematic. For example, Bypass is not an output, but the connector is named "RX_BYPASS_OUT". But you have it correct.

The other thing that can be confusing is that switching is often performed by a multiplicity of relays. For example, three relays for each of the three ANT ports. This leads to a bunch of circuit board trace stubs. Not a big deal at HF freq's, but not as easy to understand as an SP3T switch. However it has other advantages, such as the ability to ground the unused ports without a fancy, custom switch.

The receive path on the Rev. 24 basically looks like this:

Antennas-->Antenna Switching-->LPF (including bypass)-->T/R Switch-->Aux. Receive Switching (ANTs/EXTs/XVTR)-->Limiter-->Attenuator-->HPF (including bypass and 6M preamp)-->Internal PS Feeback Switch-->Bypass Switch-->RX Out

Note that the original design on the older boards used Bypass as an output so that external filters could be switched into the receive path but was re-purposed in the new design and that capability was lost.

73,

Scott
marc
Posts: 34
Joined: Mon May 15, 2017 2:40 pm

Re: Lost in 100D signal path

Postby marc » Sat Jan 20, 2018 5:37 pm

Once again, a big Thank You, Scott

Schematics of the 100D is simple... but as you said, some labeling and naming conventions are really confusing. The "outside" word around the main pcb and the different evolutions of the firmware/software/RF path was not easy to follow for a non-technical guy like me.

After building an Angelia board from component level -Gee, I learned a lot and still do- you would easily imagine I would be unable to use an "old school" bpf plugged on the J16 connector. Alex is definitively the best way to "stick" to the specs/sofware and use all the dynamic and bandwidth offered by this Hermes.

It tooks me some time before understanding most (if not all) the subtilities that the OpenHPSDR team (yourself, Phil VK6PH and many others) put in this frontend and in the underlaying code. I think I got the big picture now.

I must admit, it has been quite a lot of fun (and pure hate and rage when I was stumbling on some obscure details).

wish you all the best for the comming year (it's still time !)

Marc f6itu

Return to “Transverters”