Sky box, Pioneer DVD player, Pioneer DVD/HDD recorder, Denon AV Amp, Philips VCR, Sony CRT TV.
Scart Switcher
3 of 4
Originally Posted by Cobson:
“You couldn't post the model for the last three devices could you ? Ta!”
“You couldn't post the model for the last three devices could you ? Ta!”
Denon AV Amp: AVR-2805
I don't know the model numbers for the VCR and TV (28" Sony Wega) off-hand. They're both about 10 years old, will have to check tonight.
But they can be ignored anyway as the VCR hasn't been switched on for months I think, and I never use the TV remote control as all the pictures are supplied by external boxes and the audio goes to the amp. If you think they might be NEC protocol then I could experiment.
OK, I got a response from Shinybow. They got the idea that I was a distributor, and I guess I'll have to slowly change this view.
All salutations and fluff removed, but here is the email chain.
--------------------------------
First email:
I really hope that you can help me with a problem with one of your products.
I have a SB-5520 matrix switch and about once per day it stops responding to the Infra-Red. Searching on the web suggests that this is a common fault with this product.
See messages 12, and 43 in this thread. http://www.digitalspy.co.uk/forums/s...d.php?t=562491
I am going to have to return the product to the distributer (Skytronic in the UK) and give them a report of this problem, but I very much like the functionality of the product. Will there be a new version released with
this flaw fixed?
-----------------
First response:
Ok Can I sent you a new one to you and please be replacment to your customer after received ?
If yes please let me know your new address and phone information
TKS.
How are you to go to IFA 2007 ? We have booth at overt there..
----------------
My reply:
This seems to be a generic fault with all of these units rather than a one-off fault with one particular one. The unit seems to lock-up when receiving certain infra-red (perhaps corrupted?) from other devices, so some customers see this and some don't.
Has this fault been corrected? If not, a replacement will suffer from the same problem.
---------
Final response:
Its a wrong part.
I will be give you a report if possible you send it back to Taitwn.
---------------------------------------
I'll now muse on a suitable response! I guess the key info. we need is whether the bug is now fixed and how we tell older units from fixed ones.
Then I'll start asking about the audio amplifier issues!
Ian
All salutations and fluff removed, but here is the email chain.
--------------------------------
First email:
I really hope that you can help me with a problem with one of your products.
I have a SB-5520 matrix switch and about once per day it stops responding to the Infra-Red. Searching on the web suggests that this is a common fault with this product.
See messages 12, and 43 in this thread. http://www.digitalspy.co.uk/forums/s...d.php?t=562491
I am going to have to return the product to the distributer (Skytronic in the UK) and give them a report of this problem, but I very much like the functionality of the product. Will there be a new version released with
this flaw fixed?
-----------------
First response:
Ok Can I sent you a new one to you and please be replacment to your customer after received ?
If yes please let me know your new address and phone information
TKS.
How are you to go to IFA 2007 ? We have booth at overt there..
----------------
My reply:
This seems to be a generic fault with all of these units rather than a one-off fault with one particular one. The unit seems to lock-up when receiving certain infra-red (perhaps corrupted?) from other devices, so some customers see this and some don't.
Has this fault been corrected? If not, a replacement will suffer from the same problem.
---------
Final response:
Its a wrong part.
I will be give you a report if possible you send it back to Taitwn.
---------------------------------------
I'll now muse on a suitable response! I guess the key info. we need is whether the bug is now fixed and how we tell older units from fixed ones.
Then I'll start asking about the audio amplifier issues!
Ian
Then I asked -
----------------
Are upgraded units now available in the distribution channel? How can we tell upgraded units from older units with this Infra-red lockup issue?
----------------
And the answer was -
----------------
SB-5520
We upgraded the IR software about a one year ago. and Its working fine during this time. And we don't hear customer respone question in IR software again till Today.
I know few of older units have IR luckup problem when use
video devices connect to display.
Anyway the upgraded unit is work fine..
----------------
So, I may try and get the retailer I bought from to do a swap-out. But first I'll hunt for a serial number and see if the chap at Shinybow has any tracking of when the fix was done.
Ian
----------------
Are upgraded units now available in the distribution channel? How can we tell upgraded units from older units with this Infra-red lockup issue?
----------------
And the answer was -
----------------
SB-5520
We upgraded the IR software about a one year ago. and Its working fine during this time. And we don't hear customer respone question in IR software again till Today.
I know few of older units have IR luckup problem when use
video devices connect to display.
Anyway the upgraded unit is work fine..
----------------
So, I may try and get the retailer I bought from to do a swap-out. But first I'll hunt for a serial number and see if the chap at Shinybow has any tracking of when the fix was done.
Ian
Very strange! Particularly as I have e-mailed them about this fault months ago and e-mailed them again a couple of days ago.
Don't suppose you would be willing to give the e-mail address of your contact (unless you think I might upset your delicate negotiations!)
Don't suppose you would be willing to give the e-mail address of your contact (unless you think I might upset your delicate negotiations!)
It's just sales@shinybow.com.tw
I think that you resending your previous message and adding this address as a CC is fine - no need to reference my discussion.
The only delicate bit really is me working out how to get a unit that doesn't lock up! A mixture of good/bad units in the channel would explain what people are seeing.
Tomorrow, I'm going to search my unit for a date code and/or serial number.
Ian
I think that you resending your previous message and adding this address as a CC is fine - no need to reference my discussion.
The only delicate bit really is me working out how to get a unit that doesn't lock up! A mixture of good/bad units in the channel would explain what people are seeing.
Tomorrow, I'm going to search my unit for a date code and/or serial number.
Ian
OK, I've also now fixed the farty audio. The amp really isn't up to driving any kind of capacitive load, which is a shame given the likely applications of this box.
A pair of 75 ohm resistors (first that came to hand of about the right value!) in the audio outputs have fixed the problem. I just added them to the SCART lead that takes the matrix output to the wall socket leading to my lounge.
I'm sure this will be losing some top end but it sounds plenty good enough to TV watching and I tested it on both a music channel (for the high notes) and some Battlestar Gallactica (for the thuds and bangs) and it performed very well.
I really could get to like this matrix switch. Hell, even if the Ir doesn't get fixed, the X10 power cycling work-around isn't intollerable. I can even arrange for it to get power cycled automatically at 5am every morning!
Ian
A pair of 75 ohm resistors (first that came to hand of about the right value!) in the audio outputs have fixed the problem. I just added them to the SCART lead that takes the matrix output to the wall socket leading to my lounge.
I'm sure this will be losing some top end but it sounds plenty good enough to TV watching and I tested it on both a music channel (for the high notes) and some Battlestar Gallactica (for the thuds and bangs) and it performed very well.
I really could get to like this matrix switch. Hell, even if the Ir doesn't get fixed, the X10 power cycling work-around isn't intollerable. I can even arrange for it to get power cycled automatically at 5am every morning!
Ian
Originally Posted by gadgetmind:
“A pair of 75 ohm resistors (first that came to hand of about the right value!) in the audio outputs have fixed the problem. I just added them to the SCART lead that takes the matrix output to the wall socket leading to my lounge.”
“A pair of 75 ohm resistors (first that came to hand of about the right value!) in the audio outputs have fixed the problem. I just added them to the SCART lead that takes the matrix output to the wall socket leading to my lounge.”
How have you done that? Opened up the plug, cut the audio wires, and bridged the cut with the resistors?
Quote:
“I really could get to like this matrix switch...”
“I really could get to like this matrix switch...”
Yep, I like it too! Just wondering, have you seen the same problem I have with reduced picture quality when both outputs are set to the same source?
BTW I have the 6-way version, people who've experienced the IR lockup have the 4-way version, maybe the production runs/batches were different.
Originally Posted by Chris Simon:
“BTW I have the 6-way version, people who've experienced the IR lockup have the 4-way version, maybe the production runs/batches were different.”
“BTW I have the 6-way version, people who've experienced the IR lockup have the 4-way version, maybe the production runs/batches were different.”
Looking at your devices, From what I can dig up from the JP1 and Harmony remote databases I am pretty certain that none of them use the NEC remote control protocol. I think this is the reason why you don't have the IR lockup. If you were to get any devices that do use this protocol (such as a Topfield PVR) then I suspect you would start to see this problem.
Do you have a programmable remote ? If so you could load up the toppy device code and spend a few happy hours zapping your switch to see if it locks up!
Gatgetmind: I don't suppose you could post your devices ? If we find one that uses NEC that that would pretty much confirm the problem.
I got a further response from my contact:
"Hi Chris,
We could not duplicate the issue, as we tested some units a few weeks ago, but it is on our engineering list for things to do for next 2 month. So i hope we will get to the bottom of it, if it still has not been fixed.
Thank you"
I suspect that the IR issue that they fixed a year ago is different to the one that we are now seeing, and that the device you have is from current stock. I just don't think that Shinybow have picked up on this issue yet.
Originally Posted by Chris Simon:
“How have you done that? Opened up the plug, cut the audio wires, and bridged the cut with the resistors?”
“How have you done that? Opened up the plug, cut the audio wires, and bridged the cut with the resistors?”
Yes, exactly that. Pins 2 and 6, IIRC, as long as you're on the "far" plug rather than the one plugged into the matrix.
Quote:
“Yep, I like it too! Just wondering, have you seen the same problem I have with reduced picture quality when both outputs are set to the same source?”
“Yep, I like it too! Just wondering, have you seen the same problem I have with reduced picture quality when both outputs are set to the same source?”
No, and I checked for this pretty carefully.
Quote:
“BTW I have the 6-way version, people who've experienced the IR lockup have the 4-way version, maybe the production runs/batches were different.”
“BTW I have the 6-way version, people who've experienced the IR lockup have the 4-way version, maybe the production runs/batches were different.”
Hell, I'd pay extra for the 6-way if the thing would stop locking up!
Ian
Originally Posted by Cobson:
“Do you have a programmable remote ?”
“Do you have a programmable remote ?”
No, afraid not. I just have 7 normal ones...
Originally Posted by Cobson:
“Gatgetmind: I don't suppose you could post your devices ? If we find one that uses NEC that that would pretty much confirm the problem.”
“Gatgetmind: I don't suppose you could post your devices ? If we find one that uses NEC that that would pretty much confirm the problem.”
OK, here goes. Long list as I use Ir senders.
Toshiba 20WLT56 TV (rebadged Vestel?)
Humax 9200T PVR
Samsung LE26R74 TV
Ir->X10 bridge
SoundGraph iMon Inside PC remote
Goodmans GDB5 Freeview box (rarely used)
Roku Soundbridge M1000 networked music player
LG DR175 DVD recorder
Pure DMX50 DAB Mini system
The top two are those that are used the most.
I can get the pulse on/off times from the remote database, draw a quick graph, compare with published protocols (NEC, RC5, etc.) to see what things are using, but it's a wee bit of work.
Ian
Hmmmm, I was just firing various Ir codes at it, and it flickers its light for pretty much everything. Then it "rolled its eyes" - it did an odd rotating effect with both 7-segment displays.
I then started firing its own remote at it, but bouncing the signal off walls, masking it, and hitting the buttons like crazy. I got it to do the "eye rolling" again and then got it to lock up! So, it will lock if it gets dodgy Ir signals from its own remote.
Sweet!
Ian
I then started firing its own remote at it, but bouncing the signal off walls, masking it, and hitting the buttons like crazy. I got it to do the "eye rolling" again and then got it to lock up! So, it will lock if it gets dodgy Ir signals from its own remote.
Sweet!
Ian
The Humax uses the NEC protocol, and so does just about every Toshiba TV so I reckon that they are the problem.
All the evidence points to the switch not being able to filter out NEC protocol signals from other devices (including its own mangled signals).
All the evidence points to the switch not being able to filter out NEC protocol signals from other devices (including its own mangled signals).
Originally Posted by Cobson:
“The Humax uses the NEC protocol, and so does just about every Toshiba TV so I reckon that they are the problem.
All the evidence points to the switch not being able to filter out NEC protocol signals from other devices (including its own mangled signals).”
“The Humax uses the NEC protocol, and so does just about every Toshiba TV so I reckon that they are the problem.
All the evidence points to the switch not being able to filter out NEC protocol signals from other devices (including its own mangled signals).”
I tend to agree.
However, the stuff I got from shinybow suggests that they did have a problem but reckon they fixed it.
So -
1) This is a slightly different problem.
2) They didn't get the fix into all products.
3) A manufacturing SNAFU means an old version of the code got into some units.
4) There is some old product in the channel.
5) Something else.
The request that I send my unit back to Taiwan suggests that they suspect some combination of the above!
I think I'll come clean about my end-user status and see what they suggest.
Ian
The guy that I have been talking to is from Lenexpo who are Shinybow's US distributors (mkhain@lenexpo-electronics.com). It might be worth you dropping him a message about your own findings as he seems willing to investigate the issue. Here is the latest from him (I was asking whether the fault described on this forum is the one that was fixed a year ago. Personally I suspect it is a different fault.):
"Chris,
We are not %100 sure whether the issue had been fixed fully, i know that we start using different sensors about a year ago on remote controls. We just could not duplicate the issue. It might be the issue with old stock or unit is still not fully repaired, i will send another message to our engineering dep and hope they can address it sooner.
Thank you"
"Chris,
We are not %100 sure whether the issue had been fixed fully, i know that we start using different sensors about a year ago on remote controls. We just could not duplicate the issue. It might be the issue with old stock or unit is still not fully repaired, i will send another message to our engineering dep and hope they can address it sooner.
Thank you"
OK, I'm busy today, but then I'll send a message. But I'm in the UK.
Also, I didn't buy from Skytronic, who are the UK distributor, but from someone who had everything I needed in stock and who were slightly cheaper. This frugality might backfire but Skytronic seemed friendly so I'll bang them an email.
I'll also email the place I bought from, but they sell a very wide range of kit and will lack specialist knowledge.
Ian
Also, I didn't buy from Skytronic, who are the UK distributor, but from someone who had everything I needed in stock and who were slightly cheaper. This frugality might backfire but Skytronic seemed friendly so I'll bang them an email.
I'll also email the place I bought from, but they sell a very wide range of kit and will lack specialist knowledge.
Ian
I've had a squint at the Ir codes, and both my Toshiba TV and my PVR use the NEC protocol.
This matrix switch appears to require address 10h. The commands are all over the place. The lower nibble is F for output 1 and 7 for output 2. I can see little pattern in the other four bits, but only have 6 of the 9 codes written down, and may have made the odd mistake going from Ir on/off times to bit patterns.
Ian
This matrix switch appears to require address 10h. The commands are all over the place. The lower nibble is F for output 1 and 7 for output 2. I can see little pattern in the other four bits, but only have 6 of the 9 codes written down, and may have made the odd mistake going from Ir on/off times to bit patterns.
Ian
Do you have one of the One For All programmable remotes ? If so, a good resource is the JP1 hacking project at http://www.hifi-remote.com/jp1/index.shtml. In particular, you can use the OFA remote to learn the signals from any other remote and then run the ir.exe program to analyse the learnt signals and tell you what protocol is in use, what device code(s) are being used, and what the byte value of each signal is.
I used to have a list of the codes for this switch but I don't seem to have kept it. I do remember that it uses device code zero though.
I used to have a list of the codes for this switch but I don't seem to have kept it. I do remember that it uses device code zero though.
I have a remote called an MX500 and a bit of hardware called an Ir clone. The MX500s can clone either one device or the whole config to another remote, so a chap reverse engineered the protocol and you can now view/edit the config on a PC.
I'll have another squint at the codes as going from the on/off times to command/address can be error prone, though NEC protocol is one of the easiest.
I recently had to hand-craft the pulse times for a SoundGraph iMon Inside. They use a custom 32-bit data burst, with no error checking, and seemingly no reclocking. My remote failed to learn the codes, so I wrote some Python to generate the on/off times and typed them in hy hand! The protocol is so lame that my four remotes all need slightly different data rates - they only allow 1% clock drift before the end of the packet gets corrupted.
[later]
OK, here are the on/off times with commentary.
340, 170 - AGC calibration and start bit
21, 21 - Address 10h
21, 21
21, 21
21, 64
21, 21
21, 21
21, 21
21, 21
21, 64 - Address inverted
21, 64
21, 64
21, 21
21, 64
21, 64
21, 64
21, 64
21, 64
21, 21 - Comand 28h
21, 21
21, 64
21, 21
21, 64
21, 21
21, 21
21, 21
21, 64 - Command inverted
21, 64
21, 21
21, 64
21, 21
21, 64
21, 64
21, 64
21, 1511 - End bit and gap
All typed by hand, so maybe the odd error!
Ian
I'll have another squint at the codes as going from the on/off times to command/address can be error prone, though NEC protocol is one of the easiest.
I recently had to hand-craft the pulse times for a SoundGraph iMon Inside. They use a custom 32-bit data burst, with no error checking, and seemingly no reclocking. My remote failed to learn the codes, so I wrote some Python to generate the on/off times and typed them in hy hand! The protocol is so lame that my four remotes all need slightly different data rates - they only allow 1% clock drift before the end of the packet gets corrupted.
[later]
OK, here are the on/off times with commentary.
340, 170 - AGC calibration and start bit
21, 21 - Address 10h
21, 21
21, 21
21, 64
21, 21
21, 21
21, 21
21, 21
21, 64 - Address inverted
21, 64
21, 64
21, 21
21, 64
21, 64
21, 64
21, 64
21, 64
21, 21 - Comand 28h
21, 21
21, 64
21, 21
21, 64
21, 21
21, 21
21, 21
21, 64 - Command inverted
21, 64
21, 21
21, 64
21, 21
21, 64
21, 64
21, 64
21, 1511 - End bit and gap
All typed by hand, so maybe the odd error!
Ian
BTW, this is the NEC protocol description that I used.
http://www.sbprojects.com/knowledge/ir/nec.htm
There is *nothing* in the NEC protocol that should even phase even the most basic of embedded uPs. It's just lame coding IMO.
If this thing keeps locking up, and if the manufacturers don't find a fix, I may open it up and have a scout around. I don't reckon that it would be too hard to simply replace its uP with an AVR or a PIC. I have already built a few AVR projects that handle a variety of Ir so all I'd need to add is the code to drive whatever system they used to drive the 7-segment displays and the SCART switching elements.
Of course, if Shinybow would simply let us have the details of the uP and a copy of the code (hell, even binary as I can disassemble it!) or if they haven't blown the security fuse ...
Ian
http://www.sbprojects.com/knowledge/ir/nec.htm
There is *nothing* in the NEC protocol that should even phase even the most basic of embedded uPs. It's just lame coding IMO.
If this thing keeps locking up, and if the manufacturers don't find a fix, I may open it up and have a scout around. I don't reckon that it would be too hard to simply replace its uP with an AVR or a PIC. I have already built a few AVR projects that handle a variety of Ir so all I'd need to add is the code to drive whatever system they used to drive the 7-segment displays and the SCART switching elements.
Of course, if Shinybow would simply let us have the details of the uP and a copy of the code (hell, even binary as I can disassemble it!) or if they haven't blown the security fuse ...
Ian
Hmmm, replacing the Ir processor may be easier than I thought.
The box just locked up again with the display saying "1 2"
I used the front-panel to change to "2 2" - the SCART also switched.
I then power cycled, and the setting went back to "1 2"
So I suspect there are two processors/chips in the system.
One handles the display, front panel buttons, and the SCART switching.
The other handles Ir reception, setting storage, and sends the setting change commands to the first processor (serial?, I2C?, parallel?, 3-wire?, whatever!)
I'll pass this information on to Shinybow.
Ian
The box just locked up again with the display saying "1 2"
I used the front-panel to change to "2 2" - the SCART also switched.
I then power cycled, and the setting went back to "1 2"
So I suspect there are two processors/chips in the system.
One handles the display, front panel buttons, and the SCART switching.
The other handles Ir reception, setting storage, and sends the setting change commands to the first processor (serial?, I2C?, parallel?, 3-wire?, whatever!)
I'll pass this information on to Shinybow.
Ian
3 of 4
