Moderator Control Panel ]

TBS6522 unicable support

Re: TBS6522 unicable support

Postby oldgringo » Mon Nov 05, 2018 3:47 am

Hello,

It is useless to prepare teamviewer because I have this computer in the cottage without the reach of internet and with weak terrestrial tv signal. That's why I have chosen sat card for this desktop. And that is also the reason I can read the forum only at specific time.

I tried all existing software for sat viewing cards on Windows platform but none of them can tune this card in unicable II mode. There is probably a small difference between unicable II timing for multiswich and for direct connected LNBs (of course with necessary power inserters, combiners and splitters). I use two most advanced applications in this area - dvbviewer and dvbdream (latest TBS version). On both I can tune unicable I band (with the limit of 2 sat positions - simple diseqc) but no unicable II. Even the same band which is assigned to both unicable I+II system which works on unicable I stops immediately working once I change the system to unicable II (on dvbviewer, dvbdream cannot use user bands from unicable II bands 9-32). My limitation is that I can use only 3 user bands from unicable II range. One of them is working properly on sat receiver and the another two are intended for this sat card. The quality of the signal is 90-100% (I saw it when we tested unicable I user band) so this won't be the source of the issue. I would be thankful for any instruction which I could test on my own.
Thanks

PS: I found the mail in the spam folder. I am sorry but my mobile mail application doesn't notice me about such messages.
oldgringo
 
Posts: 13
Joined: Thu Sep 20, 2018 2:35 pm

Re: TBS6522 unicable support

Postby oldgringo » Fri Dec 28, 2018 8:04 am

I have tested the diseqc timing for unicable LNB's on digital scope and I found out that there is something strange in the proces of switching. I have attached the files with scope pictures for both unicable I and II setup. It is visible that there is something wrong with 22kHz signal which shouldn't be the part of the signal as a "pilot signal" but only as a diseqc carrier. It looks like the 22kHz generator starts sooner and finishes later than it should. And maybe just due this (but not necessarily, I guess that unicable II needs a bit more time just for switch setup because all unicable II lnb's contain built-in switch), the diseqc command doesn't get into the lnb at the specific time when the voltage is up on 18V. I found out that only first 28 bits (including parity bits) gets on time and the rest doesn't reach lnb in 18V state. Could it be somehow tweaked in the driver? It would also be nice to clean up the 22kHz remains in the signal to correct the shape (simple voltage rise with diseqc sequence on top without any other signal), it will help the stability and speed. If you need help with tests I can test it. Thanks.
Attachments
TBS6522 Unicable 2.jpg
Timing for unicable II
TBS6522 Unicable 1.jpg
Timing for unicable I
oldgringo
 
Posts: 13
Joined: Thu Sep 20, 2018 2:35 pm

Re: TBS6522 unicable support

Postby daowei » Sat Dec 29, 2018 11:04 am

oldgringo Wrote:I have tested the diseqc timing for unicable LNB's on digital scope and I found out that there is something strange in the proces of switching. I have attached the files with scope pictures for both unicable I and II setup. It is visible that there is something wrong with 22kHz signal which shouldn't be the part of the signal as a "pilot signal" but only as a diseqc carrier. It looks like the 22kHz generator starts sooner and finishes later than it should. And maybe just due this (but not necessarily, I guess that unicable II needs a bit more time just for switch setup because all unicable II lnb's contain built-in switch), the diseqc command doesn't get into the lnb at the specific time when the voltage is up on 18V. I found out that only first 28 bits (including parity bits) gets on time and the rest doesn't reach lnb in 18V state. Could it be somehow tweaked in the driver? It would also be nice to clean up the 22kHz remains in the signal to correct the shape (simple voltage rise with diseqc sequence on top without any other signal), it will help the stability and speed. If you need help with tests I can test it. Thanks.


Hi,
Thanks for your detailed testing report.
Currently, we do not have a "unicable ' II" to test. We need to learn more about the "unicable II", and find out what's the difference/special.
Can you tell us which kind of "unicable II" you're using? can send us a link or tell us the model name?
Maybe you can use "unicable I" or "Dish----to the tuner card directly" first.

Thanks
daowei
 
Posts: 137
Joined: Fri Aug 12, 2016 6:17 pm

Re: TBS6522 unicable support

Postby oldgringo » Sun Dec 30, 2018 6:23 am

I have no other option but only unicable II LNBs (provided by my neighbor in the cottage). These LNBs are able to work like unicable I device but as you can see in my previous post there are common issues. My LNBs are 4x Inverto IDLU-32UL40-UNBOO-OPP connected through unicable combiners into one block which is able (similar to 2x Inverto IDLU-UWT110-CUO1O-32P switches with wideband LNBs) tune 4 different satellite positions. It is possible to connect 32 independent tuners to this LNB block (it contains 32 user bands, each for one physical/virtual tuner) and I have assigned 3 of them. That's in short my current situation.

As I wrote earlier I have now the possibility to use digital scope to see what signal has been sent into the LNB device from different software from 2 different PCIE cards. I am in touch with programmer of this software to test more options directly with their product (like private beta version) but we would need more information about the driver's behavior. Especially what is the exact order and timing of the instruction during tune process. I would need the most detailed description of what is expected to send to the driver and the proper reaction of the driver to these inputs in full order. Something like "send command for set voltage up to 18V to driver, wait 15ms (or until driver return something), send proper diseqc command to the driver, wait 15ms (or wait for the reaction of the driver), send command for reset voltage down to 13V and wait another 15ms" or something similar. What we are curious the most are the reactions of the driver (what operations it executes as a reaction on each specific instruction, i.e. control of the 22kHz generator etc.), some error reactions which can lead to drop of the voltage back to 13V before original instruction for it has been accepted or something similar. We can try to tune software or even to give you a suggestion for driver tweaks. It is also possible that we can then tweak existing software without any additional requirements for driver development, but we must know its exact function. We were already able to make small improvements but it is still quite far from ideal.

If you don't want to publish such information here on the forum you can send it directly to my email and I will provide it to programmer. I guess it's a win-win deal for all - for you, for dvb software developer and also for me. It also might be useful to have something like "driver's api description" to publish for developers.
Thank you in advance for each information.
oldgringo
 
Posts: 13
Joined: Thu Sep 20, 2018 2:35 pm

Re: TBS6522 unicable support

Postby steven » Fri Jan 04, 2019 10:11 am

Hi oldgringo

Thanks so much for your detail feedbakc ,can you take some picture about your DVBDream unicable 2 setting picture ?
we will check at our side with this.
Thanks again for your cooperation

kind regards

steven
steven
 
Posts: 2239
Joined: Fri Aug 06, 2010 3:23 pm

Re: TBS6522 unicable support

Postby rel » Wed Jan 09, 2019 5:21 pm

Hi oldringo,

Here is the latest DVB Dream OEM version (to test) with Unicable 2 support.
Please let me know if things work for you, it supports up to 32 slots (userbands) now, when you select "Unicable II" LNB type instead of "Unicable"

https://my.pcloud.com/publink/show?code ... gt9ucOlpXX

Scope timing related issue should not be a problem on DVB Dream but let me know if there is any.
rel
 
Posts: 2
Joined: Tue Jan 08, 2019 8:23 pm

Re: TBS6522 unicable support

Postby oldgringo » Sun Jan 13, 2019 6:57 am

It is almost fine from the setup point of view. There are some incorrect parts which I will try to describe now.

1/ The setup section for unicable II sat position (it's called unicable chain) should be a separate option in LNB section (like a PIN for user band). The reason is that once I set the DiseqC switch 1.0/2.0 for physical switch then I get the control command for such type of equipment (classic command E0 XX XX XX XX). It is fine because there may be actually a physical DiseqC switch where whole unicable II equipment (LNB's/switch) is connected. So far so good. But in case of unicable II LNB's/switch there are all sats on the same "port" (no physical switch needed, see 1st attachment) in the setup and each sat position must have specific unicable chain (A, B, C or D) assigned. These chains are later (in tuning phase) translated to
A = POS A/OPT A
B = POS B/OPT A
C = POS A/OPT B
D = POS B/OPT B
data part included in 7X XX XX XX command. Each sat position in the treelist should contain its own POS/OPT parameter to properly tune one and only LNB (specific unicable chain A-D). In current setup I can tune only LNB from chain A (default one), the rest of them does not get correct signal for tuning.

2/ In case of change polarization from H to V or vice versa I can see 2 DiseqC commands (I use DC in next text) which is incorrect (see 2nt attachment). The first of them is E0 XX XX XX XX and the second unicable II command 71 XX XX XX. In case of unicable II there is a command for internal switch directly in 7X XX XX XX and therefore first DC is not needed (shouldn't be there). It even comes during 13V period so it cannot be accepted anyway. By specification there might be only one DC in each 18V time period. This part is crucial.
In case of tuning of the frequency with the same polarization everything is correct (3rd attachment).

3/ The time interval for 18V should be 50ms longer otherwise 18V pulse sometimes ends before complete DC has been sent to LNB.

4/ I miss DC for freeing used user band during quitting of the application (in case of pin assignment the user band stays "locked" so it should be freed by specific DC). There should a be simple test whether each active tuner is PIN protected and in positive reply its user band should be freed (by sending 71 XX XX XX command for this user band where frequency is set to 0).

These are my findings so far. Thanks for the improvements.
Attachments
Unicable_II_LNBs.jpg
Unicable II LNB assignment
DiseqC1.jpg
Dirrerent polarity
DiseqC2.jpg
The same polarity
oldgringo
 
Posts: 13
Joined: Thu Sep 20, 2018 2:35 pm

Re: TBS6522 unicable support

Postby steven » Wed Jan 16, 2019 10:37 am

Hi oldgringo

Thanks so much for your detail feedback

Kind Regards

steven
steven
 
Posts: 2239
Joined: Fri Aug 06, 2010 3:23 pm

Re: TBS6522 unicable support

Postby rel » Tue Jan 22, 2019 5:47 pm

oldgringo Wrote:
4/ I miss DC for freeing used user band during quitting of the application (in case of pin assignment the user band stays "locked" so it should be freed by specific DC). There should a be simple test whether each active tuner is PIN protected and in positive reply its user band should be freed (by sending 71 XX XX XX command for this user band where frequency is set to 0).

These are my findings so far. Thanks for the improvements.


https://my.pcloud.com/publink/show?code ... u2FmG2Rmok

Here is the new OEM version with fixes.
I think I have fixed all of the points. But not sure the last one works, it justs tries to set unicable frequency to 0 with the given pin code and slot number, while quitting .

"Sw. Pos" now allows unicable internal switch to be specified. (2 ports for Unicable 1 and 4 ports for Unicable 2)
You need to select Diseqc switch type as None at the top while adding satellites to not get any diseqc command other than unicable commands. (if you have a diseqc switch infront of the unicable device, then you can set it to related diseqc type, no prob.)

Thanks for the detailed info.
rel
 
Posts: 2
Joined: Tue Jan 08, 2019 8:23 pm

Re: TBS6522 unicable support

Postby oldgringo » Thu Jan 31, 2019 6:30 am

It looks good, everything works like it should (from first quick sight). I'm going to test it deeper (with digital scope) during the weekend. Anyway, great job, thanks! ;)
oldgringo
 
Posts: 13
Joined: Thu Sep 20, 2018 2:35 pm

PreviousNext

Return to TBS6522 Multi-standard Dual Tuner PCI-e Card

Who is online

Users browsing this forum: No registered users and 5 guests