Hauppauge WinTV-NOVA-TD-Stick
A DVB-T USB device from Hauppauge.
It is supported under Linux.
Similar models
Elgato EyeTV diversity is a clone of this usb stick
Overview/Features
This device features two tuners and is supplied with two antennas. You'll have adapter0 and adapter1 in /dev/dvb, which you can use separately. Adapter0 seems to be the connector on the side of the stick? Note that you need to have 2 antenna leads plugged into the stick, one in the end, and the other into the side in order to get both tuners to work! Owner's manual is here.
The "Diversity" option is a hardware based feature that allows for the device's two receivers to be configured in a combined use mode to achieve better reception on a single channel. The diversity feature of the DiBcom demodulators is currently not implemented in the Linux-DVB drivers, so only the dual tuner configuration is presently supported on such devices [1].
This problem seems to be fixed now- see this information. --Indulis 09:39, 24 July 2008 (CEST)
Components Used
- 2× Microtune MT2266 tuners
- Dibcom DiB7700P DVB-T demodulator (DiB7070P? no such chip as 7700P!)
Identification
This device's USB ID is 2040:9580.
Additional information: I have bought a WinTV-NOVA-TD-Stick that has a different USB ID: 2040:5200. Dirks 20:05, 27 October 2008 (CET)
Making it Work (generic for all dib0700)
Firmware
August 21, 2008 - New firmware file fixing the last cause for i2c errors and disconnects and providing a new, more modular i2c request formatting.
You will need the dvb-usb-dib0700-1.20.fw firmware file in /lib/firmware or the relevant place for your distribution.
You may need to change the name of the file to dvb-usb-dib0700-1.10.fw or create a link until the driver code reflects that change.
For archival purposes: dvb-usb-dib0700-1.10.fw firmware file
August 29,2008 - Issues with Firmware 1.20. Some issues have been found with the latest version of the firmware. Users may wish to continue to use 1.10 unless they have patched their v4l-dvb code with dib0700_new_i2c_api.patch.
November 15,2008 - Issues with Firmware 1.20.
- The above mentioned dib0700_new_12c_api.patch is not available discretely but is now rolled into the mercurial drivers
dvb-usb-dib0700-1.20.fw firmware fileis now stable for reception, but remote control functionality is broken; any key press is repeated until the next key is pressed. The only way to get remote control functionality presently is to roll back to 1.10 firmware and suffer the occasional disconnect.- The mercurial drivers have been changed so they now load 1.20 firmware. To revert to 1.10 firmware you need to rename your firmware file to dvb-usb-dib0700-1.20.fw or provide a link of that name.
- To avoid spurious remote control signals with 1.20 firmware, you need to edit /etc/modprobe.d/options or from Ubuntu onwards /etc/modprobe.d/options.confand add:
options dvb_usb disable-rc-polling=1
November 28,2008 - i2c errors. Changes were made to the remote control drivers on November 16,2008 to correct the repeat key problem. The card is generally stable for dual tuner reception and remote control function with Firmware 1.20.
November 10,2009 - mt2060 I2C write failed. Possible regression of a driver bug raised against Ubuntu running 2.6.27-14 and 2.6.31-2.17 causing mt2060 I2C errors in MythTV useage with firmware 1.20. https://bugs.launchpad.net/ubuntu/+source/linux/+bug/397696 Recommend check the kernel extensions listed here for Low Noise Activation and rc_polling are loaded with correct config file name for your distribution, EIT listings information is turned off until a suitable delay (500ms-1000ms)is added to a single card (not both) and the card has correctly been added to the database as two tuners (no additional NULL entries) in the mythtv recordcard table.
Drivers
It requires the dib0700 driver. Just use Mercurial by following the How to Obtain, Build and Install V4L-DVB Device Drivers instructions.
Forcing the activation of LNAs (Low Noise Amplifier)
You may have to force LNA to get this card working:
In /etc/modprobe.d/options add:
options dvb_usb_dib0700 force_lna_activation=1
Disabling the remote control sensor
You may want to disable the remote control sensor if you are using another one and want to avoid error messages in the logs:
In /etc/modprobe.d/options add:
options dvb_usb disable_rc_polling=1
All relevant kernel modules options
In /etc/modprobe.d/options add:
options [module name] [option name]=[setting]
Get the parameters list using
modinfo [name of kernel module]
The debug values are bit fields, with each bit representing a different category. Add values to turn on multiple debugging categories.
dib3000mc
- debug
- Turn on debugging
- Values: integer
- Default: 0 (off)
- buggy_sfn_workaround
- Enable work-around for buggy SFNs
- Values: integer
- Default: 0 (disabled)
mt2060
- debug
- Turn on debugging
- Values: integer
- Default: 0 (off)
dvb_usb_dib0700
- force_lna_activation
- Force the activation of LNAs (Low Noise Amplifier), if applicable for the device
- Values: integer
- Default: 0 (automatic/off)
- dvb_usb_dib0700_ir_proto
- Set IR protocol
- Values: integer 0=NEC, 1=RC5, 2=RC6
- Default: 1
- debug
- Set debugging level
- Values: integer (bitmap) 1=info, 2=fw, 4=fwdata, 8=data
- Default: 0 (none)
dvb_usb
- debug
- Set debugging level
- Values: integer (bitmap) 1=info, 2=xfer, 4=pll, 8=ts, 16=err, 32=rc, 64=fw, 128=mem, 256=uxfer
- Default: 0 (none)
- disable_rc_polling
- Disable remote control polling
- Values: integer
- Default: 0 (enabled)
- force_pid_filter_usage
- Force all DVB USB devices to use a PID filter, if any
- Values: integer
- Default: 0 (disabled)
dvb_core
- dvb_net_debug
- Enable debug messages
- Values: integer
- Default: 0 (disabled)
- frontend_debug
- Turn on frontend core debugging
- Values: integer
- Default: 0 (off)
- dvb_shutdown_timeout
- Wait n seconds after close() before suspending hardware
- Values: integer
- Default: 0
- dvb_force_auto_inversion
- Set whether INVERSION_AUTO is forced on
- Values: integer
- Default: 0 (off)
- dvb_override_tune_delay
- Wait n milliseconds for lock after a tuning attempt
- Values: integer
- Default: 0
- dvb_powerdown_on_sleep
- Turn LNB power off on sleep
- Values: integer
- Default: 1 (enabled)
- cam_debug
- Enable verbose debug messages
- Values: integer
- Default: 0 (off)
- debug
- Turn on debugging
- Values: integer
- Default: 0
- dvbdev_debug
- Turn on device debugging
- Values: integer
- Default: 0 (off)
dibx000_common
- debug
- Turn on debugging
- Values: integer
- Default: 0 (off)
Remote control support
Using evdev
As long as the evdev module is loaded, a remote that is recogniced as hid device will be treated as a usb keyboard and this means that you can avoid using lirc.
However, many of the keys on your remote may generate keycodes which are not mapped to anything, by default.
In X you can use xev to find the keycodes and xmodmap to map them to useful symbols. Unfortunately, some keys may generate keycodes that X doesn't recognize at all and the device does not support keymaps, or this would be easy to fix.
Using LIRC
Usually remote controls in linux are managed by the lirc software collection.
To get lirc up and running you need to configure some things.
- Settings for the hardware
- Where does lirc get its input from? aka. the DEVICE. E.g. /dev/input/event3
- How to handle the input? aka. the DRIVER. E.g. devinput
- Settings for mapping driver output generated by your remote (a bunch of hex numbers) to key names (something like 0..9, Volume+, Next, Record)
- Settings for mapping key presses to actions (usually located in your .lircrc)
Mythubuntu case
On mythubuntu 10.10, you just have to add this line in /etc/udev/rules.d/65-persistent-hauppauge.rules
SUBSYSTEM=="input", KERNEL=="event*", ATTRS{idVendor}=="2040", ATTRS{idProduct}=="8400", SYMLINK+="lirc0"
Device/driver settings
Find the IR receiver's device by looking in the dmesg output for a line similar to:
input: IR-receiver inside an USB DVB receiver as /class/input/input4
Additionally, the IR receiver will be listed if you execute the command:
cat /proc/bus/input/devices
For example:
I: Bus=0003 Vendor=2040 Product=9950 Version=0100 N: Name="IR-receiver inside an USB DVB receiver" P: Phys=usb-0000:07:01.2-1/ir0 S: Sysfs=/class/input/input4 U: Uniq= H: Handlers=kbd event4 B: EV=3 B: KEY=14afc336 284284d00000000 0 480058000 219040000801 9e96c000000000 90020000000ffd
In this example, the remote control gives output into /dev/input/event4.
The event number depends on your particular system and can vary.
Eventually this event number can even vary at every reboot.
You could create a new udev rule in /etc/udev/rules.d/65-persistent-hauppauge.rules.
KERNEL=="event*", ATTRS{name}=="IR-receiver inside an USB DVB receiver", SYMLINK+="input/dvb-ir"
This would make IR receivers handled by the usb_dvb framework always always be linked to /dev/input/dvb-ir.
But Linux systems running recent udev will automatically create non-varying names, a nicer and automatic way of providing a stable input event name:
$ ls -la /dev/input/by-path/ total 0 drwxr-xr-x 2 root root 140 2008-02-07 20:31 . drwxr-xr-x 4 root root 280 2008-02-07 20:31 .. lrwxrwxrwx 1 root root 9 2008-02-07 20:31 pci-0000:00:1a.1-usb-0:2:1.0-event-kbd -> ../event1 lrwxrwxrwx 1 root root 9 2008-02-07 20:31 pci-0000:00:1a.1-usb-0:2:1.1-event-mouse -> ../event2 lrwxrwxrwx 1 root root 9 2008-02-07 20:31 pci-0000:00:1a.1-usb-0:2:1.1-mouse -> ../mouse1 lrwxrwxrwx 1 root root 9 2008-02-07 20:31 pci-4-1--event-ir -> ../event4 lrwxrwxrwx 1 root root 9 2008-02-07 20:31 platform-pcspkr-event-spkr -> ../event3
LIRC will use it without needing a special kernel module. use the dev/input (or devinput. Check this with the command "lircd --device=help".) driver and specify the input event device in /etc/lirc/hardware.conf
# /etc/lirc/hardware.conf # # Arguments which will be used when launching lircd LIRCD_ARGS="" #Don't start lircmd even if there seems to be a good config file #START_LIRCMD=false #Try to load appropriate kernel modules LOAD_MODULES=true # Run "lircd --driver=help" for a list of supported drivers. DRIVER="dev/input" # If DEVICE is set to /dev/lirc and devfs is in use /dev/lirc/0 will be # automatically used instead REMOTE_DEVICE="/dev/input/by-path/pci-4-1--event-ir" MODULES="" # Default configuration files for your hardware if any LIRCD_CONF="/etc/lirc/lircd.conf" LIRCMD_CONF=""
If you have REMOTE and TRANSMITTER sections in your hardware.conf file, they should look like this:
#Chosen Remote Control REMOTE="Terratec Cinergy DT USB XS Diversity" REMOTE_MODULES="" REMOTE_DRIVER="devinput" REMOTE_DEVICE="/dev/input/by-path/pci-1-5-event-ir" REMOTE_LIRCD_CONF="/etc/lirc/lircd.conf" REMOTE_LIRCD_ARGS=""
#Chosen IR Transmitter TRANSMITTER="None" TRANSMITTER_MODULES="" TRANSMITTER_DRIVER="" TRANSMITTER_DEVICE="" TRANSMITTER_LIRCD_CONF="" TRANSMITTER_LIRCD_ARGS=""
Remote key setup
See device specific section below or try [2].
Sample .lircrc
A sample .lircrc can be found LircrcExample here.
Keys repeated twice
But there is still the problem of the key repeats for it, so that each keypress will be repeated twice. The patches, as mentioned above, may not work, but a workaround is possilbe. It is described in http://ubuntuforums.org/showthread.php?p=4253678
Simply add config = echo " > /dev/null before the main config in .mythtv/lircrc or .lircrc
begin prog = mythtv button = Mute config = echo " > /dev/null config = | ... end
So each 2nd keypress will be suppressed. This works in some application but not others (e.g. vlc).
Alternatively there is a patch for the kernel driver that solves it, it can be found here.
Finally if that doesn't work and you have the silver remote (A415-HPG-WE-A ) then changing the lircd.conf line as follows can prevent the duplicate key presses. This has the side-effect of disabling key repeats for the remote entirely. Change toggle_bit_mask 0x80000000 to toggle_bit_mask 0x00000000
Note: do not try to comment out (using #) any line in this file, or lirc won't work anymore.
Do NOT do this:
#toggle_bit_mask 0x80000000 toggle_bit_mask 0x00000000
Replace the original line instead.
Specific to the model
Specific Remote control support
If you do want to use lirc set it up to use the that device as DEVICE and devinput as DRIVER. Then try one of the config files below. If it doesn't work then you will need to use irrecord to generate a config file.
While key repeats are still buggy irrecord will fail if you obey its instructions and hold down an arbitrary key; instead, when you are told to hold an arbitrary key, press an arbitrary key repeatedly (the same one, not different ones each time) until irrecord reports that it has found the gap length.
Snowboard remote control
Grey top, black bottom, 45 buttons, snowboard shape.
Here is a proper lircd.conf:
#
# brand: Hauppauge
# model no. of remote control: 45 buttons Snowboard Shape Silver over Black
#
begin remote
name hauppauge-45-snowboard
bits 16
eps 30
aeps 100
one 0 0
zero 0 0
pre_data_bits 16
pre_data 0x1
gap 199999
toggle_bit 0
begin codes
Go 0x0162
Power 0x0074
TV 0x0179
Videos 0x0189
Music 0x0188
Pictures 0x00E2
Guide 0x016D
Radio 0x0181
ArrowUp 0x0067
ArrowLeft 0x0069
OK 0x0160
ArrowRight 0x006A
ArrowDown 0x006C
BackExit 0x009E
Menu 0x008B
VolumeUp 0x0073
VolumeDown 0x0072
PrevCh 0x016B
Mute 0x0071
ChannelUp 0x0192
ChannelDown 0x0193
Record 0x00A7
Rewind 0x00A8
SkipBack 0x0195
Play 0x00CF
Pause 0x0077
Stop 0x0080
Fwdwind 0x00D0
SkipFwd 0x0197
1 0x0002
2 0x0003
3 0x0004
4 0x0005
5 0x0006
6 0x0007
7 0x0008
8 0x0009
9 0x000A
* 0x0037
0 0x000B
# 0x0029
Red 0x018E
Green 0x018F
Yellow 0x0190
Blue 0x0191
end codes
end remote
Silver remote
Assuming, you have a silver remote, named A415-HPG-WE-A. The following keycodes should work with it, specify them in /etc/lirc/lircd.conf:
begin remote
name Hauppauge A415-HPG-WE-A
bits 16
eps 30
aeps 100
one 0 0
zero 0 0
pre_data_bits 16
pre_data 0x1
gap 403983
toggle_bit_mask 0x80000000
begin codes
TV 0x0179
Videos 0x0189
Music 0x0188
Pictures 0x00E2
Radio 0x0181
Guide 0x016D
Off 0x0074
Go 0x0162
Up 0x0067
Down 0x006C
Left 0x0069
Right 0x006A
Ok 0x0160
Menu 0x008B
Back/Exit 0x009E
Vol+ 0x0073
Vol- 0x0072
PrevCh 0x016B
Mute 0x0071
Ch+ 0x0192
Ch- 0x0193
Stop 0x0080
Record 0x00A7
Rew 0x00A8
Play 0x00CF
FFW 0x00D0
Pause 0x0077
Replay 0x0195
Skip 0x0197
1 0x0002
2 0x0003
3 0x0004
4 0x0005
5 0x0006
6 0x0007
7 0x0008
8 0x0009
9 0x000A
Star 0x0037
0 0x000B
Char 0x0029
Red 0x018E
Green 0x018F
Yellow 0x0190
Blue 0x0191
end codes
end remote
Sample kernel output
From 2.6.20-16-generic with drivers compiled from repository 4th Jan 07:
$ dmesg | grep dvb [ 40.858884] dvb-usb: found a 'Hauppauge Nova-TD Stick/Elgato Eye-TV Diversity' in cold state, will try to load a firmware [ 40.912496] dvb-usb: downloading firmware from file 'dvb-usb-dib0700-1.10.fw' [ 41.624338] dvb-usb: found a 'Hauppauge Nova-TD Stick/Elgato Eye-TV Diversity' in warm state. [ 41.624397] dvb-usb: will pass the complete MPEG2 transport stream to the software demuxer. [ 42.047026] dvb-usb: will pass the complete MPEG2 transport stream to the software demuxer. [ 42.359232] dvb-usb: schedule remote query interval to 150 msecs. [ 42.359235] dvb-usb: Hauppauge Nova-TD Stick/Elgato Eye-TV Diversity successfully initialized and connected. [ 42.359404] usbcore: registered new interface driver dvb_usb_dib0700 [ 68.220540] dvb_usb_dib0700 2-4:1.0: resuming [ 68.220812] dvb_usb_dib0700 2-4:1.0: resuming
Troubleshooting
It seems, that there exist USB-problems with mainboards using an AMD690g chipset concerning USB-disconnects. This problem is described here and on the linux-dvb Mailing list. (Disable the remote sensor?) --Doc murke 09:40, 13 March 2008 (CET)
This problem should be fixed now- see this information. --Indulis 09:39, 24 July 2008 (CEST)
Low power muxes
At least two people have indicated on the mailing list (8 April 2008 and 12 April 2008) that trouble picking up reception on low power muxes can be helped by adding an attenuator. It is thought that this may be caused by the internal amplifier taking in to account the strongest muxes only. Weakening the overall signal appears to increase the amplification of all muxes, so improving reception.
Signal power
The Nova-TD-Stick (and to a lesser degree its "brother", the Nova-T-500) have a very limited dynamic range on the input connectors. This means that if you feed the TD-Stick with an extremely strong signal, you'll see the signal strength 'stick' at 100%, but the TD-Stick will never achieve a signal lock. This typically happens when using a masthead amplifier or distribution amplifier to feed multiple TVs from a single TV aerial. Ideally the masthead amplifier should be turned down to a very low setting, and if necessary followed with a 6dB or 12dB (as necessary) attenuator. Buy a few attenuators (3dB, 6dB, 12dB and 24dB are the common versions) and mix-and-match until you get something that works. As a general rule:
- If the signal strength is less than 25%, then your aerial isn't connected (or the signal is too weak for the TD-Stick to detect).
- If the signal strength is 100% and you haven't got a channel lock (or the data appears to be corrupted -- unnamed channels, etc.), then the signal is too strong. Turn down the amplifier, or install an attenuator or two.
- Full-sized TVs have much better tuners (typically wide dynamic range superheterodyne tuners) than the Microtune silicon tuners in the TD Stick. Bear in mind that your TV can probably handle overly weak or strong signals better than your TD-Stick.
- If you know someone who owns a TV strength meter or spectrum analyser, ask to borrow it for an afternoon. A signal strength of between 60 and 75 dBuV (decibel-microvolts) measured at the antenna outlet is generally pretty close to optimum for TV receivers. Perhaps someone with a variable attenuator, a spectrum analyser and a TD-Stick could figure out the actual signal-lock range of the Microtune tuner in the TD-Stick (though this would be a largely academic exercise).
--Philpem 17:48, 10 December 2010 (UTC)
A Comment on Possible Electromagnetic Interference (jarb1 at uk2 dot net)
These devices appear to be susceptible to EM interference. This is in UK on DVB-T reception, 50Hz, 240VAC. Having two of these fed from coax Splitter into the back of a decent case (Silverstone LC04) it was found :-
- Leave the device and Co-ax feeding the device anywhere near the case power leads etc and experience screen freezes, pixelation and sound stutter.
- Plugging directly into the case on the USB port directly below the ethernet port picked up noise (Presumably from the ethernet Unshielded Twisted Pair 100M). Coaxes on this test were well routed so strongly believe this is the device.
To solve, use the USB extension leads from the box to get the actual device away from the case and run the coaxial cable well away from the rest of your cables.
Once this was done a 100% perfect picture 100% of the time.
If Hauppauge read this, please retest these devices in a controlled environment under the CE EMC directive and delete this if you can not simulate.
-- Jamesarbrown 12:01, 30 November 2008 (UTC)
Thank you for the above description! I have a newer revision of this stick, but run into the same problem. The stick didn't work on two different PCs with two different operating systems.
Then I've read your comment and connected the stick using the USB extension and used a better shielded cable and it started to work.
-- Theli 08:04, 7 December 2012 (CET)