sci.electronics.repair@googlegroups.com

Google Groups

Bram van den Heuvel <braam456987@yahoo.com>: Jul 11 02:14AM How does Windows 10 handle manual & automatic switching between Internet WiFi & Ethernet connections?   Does it automatically switch based on the best signal strength? Or does it automatically switch based on some test speed?   If I want to manually change from one connection to the other, how do I do that?   # I have a desktop with no WiFi card. # On a USB port, I have a $25 AC1200 TP Link "T4U" wifi dongle (FCCID TE7T4U, IC 8853A-T4U) https://fccid.io/TE7T4U # On the Ethernet port, I have an Ubiquiti Nanobeam M2 (802.11n) (FCCID SWX-NBM2HP, IC 6545A-NBM2HP)   My two questions are 1. How does Windows automatically decide which to use? 2. How do I manually switch from one to the other?

lifewoutmilk <lifewoutmilk@distraction.x10host.com>: Jul 10 08:27PM -0700 Bram van den Heuvel wrote on 7/10/2017 7:14 PM:   > My two questions are > 1. How does Windows automatically decide which to use? > 2. How do I manually switch from one to the other?   <https://support.microsoft.com/en-us/help/299540/an-explanation-of-the-automatic-metric-feature-for-ipv4-routes>   > The Automatic Metric feature can be useful when the routing table contains multiple routes for the same destination. For example, if you have a computer with a 10 megabit (Mb) network interface and a 100 Mb network interface, and the computer has a default gateway that is configured on both network interfaces, the Automatic Metric feature assigns a higher metric to the slower network interface. This feature can force all of the traffic that is destined for the Internet, for example, to use the fastest network interface that is available.   Basically, Windows chooses the fastest link. You can override this by adjusting the automatically assigned metric. To adjust the metric, open the adapter properties > Internet Protocol Version 4 > General > Advanced and disable Automatic metric. The adapter with the lowest value is the one Windows will choose.   If you only need a particular adapter to be used temporarily, it might be easier to either manually disconnect from wireless or disable the adapter (right-click > Disable), rather than setting the metric manually.

Bram van den Heuvel <braam456987@yahoo.com>: Jul 11 04:32AM Given <news:ok1gbd$t7i$1@dont-email.me>, lifewoutmilk > the adapter properties > Internet Protocol Version 4 > General > > Advanced and disable Automatic metric. The adapter with the lowest value > is the one Windows will choose.   Thanks to that info, I found that "automatic metric" checkbox.   Here is a picture of it on my computer desktop. https://www.turboimagehost.com/p/35488920/wifi.jpg.html   > If you only need a particular adapter to be used temporarily, it might > be easier to either manually disconnect from wireless or disable the > adapter (right-click > Disable), rather than setting the metric manually.   At this point that's exactly what I do which is I either pull the Ethernet cable out of the one wifi antenna connection at the back of the desktop or I pull the usb stick out of the usb port at the back of the desktop.   Right clicking and disabling would also work as you noted.   What I will try to figure out is how to *set* that metric so that I can just set the adapter metric to, say, 25 for the adapter I want to win and 50 for the adapter I want to lose, and when I want to switch which wifi connection it uses, I just reverse the "cost" metric.   That web page you referred me to seems great but it's complex the first two times I read it.   An explanation of the Automatic Metric feature for IPv4 routes https://support.microsoft.com/en-us/help/299540/an-explanation-of-the-automatic-metric-feature-for-ipv4-routes

Char Jackson <none@none.invalid>: Jul 11 12:56AM -0500 On Tue, 11 Jul 2017 04:32:28 +0000 (UTC), Bram van den Heuvel >just set the adapter metric to, say, 25 for the adapter I want to win and >50 for the adapter I want to lose, and when I want to switch which wifi >connection it uses, I just reverse the "cost" metric.   https://serverfault.com/questions/238695/how-can-i-set-the-metric-of-a-manually-added-route-on-windows   "route print" gets you the interface numbers, as well as the current metrics.   Plug the interface number into the following command to set the new metric. (Following command uses 25 as the interface number.)   netsh interface ipv4 set interface 25 metric=2   When you're tired of it, set it back to automatic like this:   netsh interface ipv4 set interface 25 metric=automatic   For a bit of hands-off automation, create a pair of batch files that do these tasks simply by double clicking.

Bram van den Heuvel <braam456987@yahoo.com>: Jul 11 07:46AM Given <news:9sp8mc1aph9divukbl4givtffe62kjptvh@4ax.com>, Char Jackson   > "route print" gets you the interface numbers, as well as the current > metrics.   Taking this step by step, does this output from "route print" make any sense to you?   =========================================================================== Interface List 13...c4 e9 84 a1 4c aa ......TP-LINK Wireless USB Adapter 9...c4 e9 84 a1 4c aa ......Microsoft Wi-Fi Direct Virtual Adapter 8...0a 00 27 00 00 08 ......VirtualBox Host-Only Ethernet Adapter 12...00 26 55 d9 e3 44 ......Realtek PCIe GBE Family Controller 11...00 ff 84 a3 fa 25 ......TAP-Windows Adapter V9 1...........................Software Loopback Interface 1 44...00 00 00 00 00 00 00 e0 Teredo Tunneling Pseudo-Interface 16...00 00 00 00 00 00 00 e0 Microsoft ISATAP Adapter #3 5...00 00 00 00 00 00 00 e0 Microsoft ISATAP Adapter #4 ===========================================================================   Of the 9 items in that list, I only recognize the TP-Link USB dongle, and the RealTek Ethernet adapter. I guess the VirtualBox adapter is because I have VirtualBox installed. And I guess the TAP adapter is because I sometimes use OpenVPN.   The Wi-Fi direct adapter has the same MAC address as the TP-Link dongle so it's probably inside the dongle.   But what is the "Teredo" or "ISATAP" adaptor?   Googling, both the Teredo & ISATAP things are related to IPv6 I think. https://en.wikipedia.org/wiki/Teredo_tunneling https://en.wikipedia.org/wiki/ISATAP   What does my "route print" tell you that you can tell me?

Andy Burns <usenet@andyburns.uk>: Jul 11 10:48AM +0100 Char Jackson wrote:     > netsh interface ipv4 set interface 25 metric=2   > When you're tired of it, set it back to automatic like this:   > netsh interface ipv4 set interface 25 metric=automatic   of if you want to be able to switch seamlessly from wifi to wired ethernet, create a bridge of the two NICs, and let the IP address get assigned to the bridge, rather than having IP addresses assigned to each individual NIC.

"pfjw@aol.com" <pfjw@aol.com>: Jul 11 04:03AM -0700 The Troll is back, please do not feed the troll.   Most computers - including the one I am using this minute - does this automatically and seamlessly. This troll writes for the attention.

Keith Nuttle <Keith_Nuttle@sbcglobal.net>: Jul 11 08:03AM -0400 On 7/10/2017 10:14 PM, Bram van den Heuvel wrote:   > My two questions are > 1. How does Windows automatically decide which to use? > 2. How do I manually switch from one to the other?   I am not sure exactly what you are trying to accomplish. What you are asking sort of makes sense if you have a laptop that you take different places.   On my desktop computer I have both Ethernet and wireless connections. As stated Windows 10 picks the fastest connections.   The easiest way to change the connections manually is:   Right click the MSIcon on the lower left of the tool bar;   Click Network Connections;   click Changed Adapters Options;   Enable which ever adapter you wish to use.   In three click you can change your adapter.   -- 2017: The year we lean to play the great game of Euchre

lifewoutmilk <lifewoutmilk@distraction.x10host.com>: Jul 11 06:14AM -0700 Bram van den Heuvel wrote on 7/10/2017 9:32 PM: > times I read it.   > An explanation of the Automatic Metric feature for IPv4 routes > https://support.microsoft.com/en-us/help/299540/an-explanation-of-the-automatic-metric-feature-for-ipv4-routes   <https://technet.microsoft.com/en-us/itpro/powershell/windows/nettcpip/set-netipinterface#example-2-modify-the-interface-metrics>   You could change it with a PowerShell, create two scripts, one to prefer wired, one to prefer wireless.   The PowerShell command to set the metric is:   Set-NetIPInterface -InterfaceIndex INDEX -InterfaceMetric VALUE   To get the INDEX for the adapter, run 'Get-NetAdapter' and look at the ifIndex column.

Char Jackson <none@none.invalid>: Jul 11 10:53AM -0500 On Tue, 11 Jul 2017 07:46:50 +0000 (UTC), Bram van den Heuvel >===========================================================================   >Of the 9 items in that list, I only recognize the TP-Link USB dongle, and >the RealTek Ethernet adapter.   Agreed, those are your wireless and wired interfaces, respectively, so you're interested in interface 13 (WiFi) and interface 12 (wired). Just ignore the rest for now.

John Robertson <spam@flippers.com>: Jul 10 10:20AM -0700 On 2017/07/08 7:38 PM, John-Del wrote:   >> thanks   > You can't check this diode with a dmm... I posted above:   > "I didn't read back through all the posts, but if you haven't, check the pump up diode. The anode will be connected to the Vcc and the cathode to the pump up pin of the vert IC. A weak diode here will cause all sorts of havoc (don't check it with a dmm - they often sag under load)."   That is a most interesting point, the 1N4937 diode is indeed a fast switching diode and it certainly appears to be critical for building the vertical deflection. Your point about sagging under load could indeed be correct for this problem as described by the OP.   Learn something every day, thanks!   John :-#)#

"Ian Field" <gangprobing.alien1@virginmedia.com>: Jul 10 08:17PM +0100 <mhooker32@gmail.com> wrote in message news:5dc0ef0f-2541-446c-ac84-d00d2edab8d7@googlegroups.com... > output? the same circuit( before the cap) goes to pin 3. there is only > about 1/2 volt on pin 7. is it supposed to pulse, pin 7 pulling down the > 24 v?   There's sometimes an electro with ambiguous purpose (also on audio amps). It bootstraps the output to the bias network for the output transistor bases. Its almost positive feedback that reduces drive current requirement for the output pair and improves linearity.

mhooker32@gmail.com: Jul 10 04:22PM -0700 On Monday, July 10, 2017 at 3:17:06 PM UTC-4, Ian Field wrote: > bootstraps the output to the bias network for the output transistor bases. > Its almost positive feedback that reduces drive current requirement for the > output pair and improves linearity.   just replaced the vertical deflection IC LA7851. same image, no better. regarding d360 possibly sagging, i did swap it with another on the board, and it was the same. i'll have to keep digging, but any help is appreciated.   thanks

John-Del <ohger1s@gmail.com>: Jul 10 05:06PM -0700 > just replaced the vertical deflection IC LA7851. same image, no better. regarding d360 possibly sagging, i did swap it with another on the board, and it was the same. i'll have to keep digging, but any help is appreciated.   > thanks       Not much left...   Try warming up the board with a hair dryer on high or a heat gun on low and watch the picture as you do. I'd bet dollars to donuts at this point that the picture stretches out. You said you recapped the monitor but was this all the larger electros that are known to cause issues (like a kit) or *everything*. There are often low value electrolytics back at the signal generator side used for signal coupling that get juicy on the bottom and cause issue either from low esr (they can be juicy and still check OK) or from conductive contamination from the leaked electrolyte.   If the heat doesn't expose a lazy high esr coupling cap, then try washing the board with a suitable cleaner and acid brush, removing the "jungle" IC if need be but otherwise clean between all the leads.

mhooker32@gmail.com: Jul 10 07:25PM -0700 On Monday, July 10, 2017 at 8:06:35 PM UTC-4, John-Del wrote:   > Not much left...   > Try warming up the board with a hair dryer on high or a heat gun on low and watch the picture as you do. I'd bet dollars to donuts at this point that the picture stretches out. You said you recapped the monitor but was this all the larger electros that are known to cause issues (like a kit) or *everything*. There are often low value electrolytics back at the signal generator side used for signal coupling that get juicy on the bottom and cause issue either from low esr (they can be juicy and still check OK) or from conductive contamination from the leaked electrolyte.   > If the heat doesn't expose a lazy high esr coupling cap, then try washing the board with a suitable cleaner and acid brush, removing the "jungle" IC if need be but otherwise clean between all the leads.   well, im starting to think that its going into HV shutdown, and rather than shutting down, the half scan is how it was made to shutdown. still, doesnt shutdown usually not shut down the vertical scan ? i dont think anything on this unit is the usual. the manual for it is near useless, very little info. another thing, the schematic shows two pairs of jumpers, for 13 or 19 inch use. both pairs are jumped on this board, and the same board on ebay has both pairs jumped. its a hair puller. pin 13 on the 7851 is the x ray protection pin, and it needs to be a specific voltage or the chip shuts down. the manual says that, but doesnt say the voltage. im going to end up buying a chasssi, but i'd still like to know what the heck is wrong with mine.   thanks

John-Del <ohger1s@gmail.com>: Jul 11 04:14AM -0700 > > If the heat doesn't expose a lazy high esr coupling cap, then try washing the board with a suitable cleaner and acid brush, removing the "jungle" IC if need be but otherwise clean between all the leads.   > well, im starting to think that its going into HV shutdown, and rather than shutting down, the half scan is how it was made to shutdown. still, doesnt shutdown usually not shut down the vertical scan ? i dont think anything on this unit is the usual. the manual for it is near useless, very little info. another thing, the schematic shows two pairs of jumpers, for 13 or 19 inch use. both pairs are jumped on this board, and the same board on ebay has both pairs jumped. its a hair puller. pin 13 on the 7851 is the x ray protection pin, and it needs to be a specific voltage or the chip shuts down. the manual says that, but doesnt say the voltage. im going to end up buying a chasssi, but i'd still like to know what the heck is wrong with mine.   > thanks   When projection TVs started hitting the market, manufacturers soon learned that a scan loss in either direction would immediately destroy the CRTs. In response, manufacturers learned to not start the HV until both deflections were established, and shut down same if one was lost or even diminished by a fixed percent. The point is that a vertical collapse would not be the result of a shutdown, but a cause.   Read my previous post again. One of the last CRT TVs I worked on was a Zenith 35" console with a stubborn vert (slow to fill, weird linearity issues). It turned out that the board became conductive from leaked electrolyte and dust. Heating and cooling the board affected the deflection, but no components were themselves temperature sensitive. I ended up pulling the flyback, smps transformer, and any inductor that looked like it could hold water and immersed the whole board in a parts cleaner. Once assembled, the vertical was perfect and didn't drift.

mhooker32@gmail.com: Jul 11 06:39AM -0700 On Tuesday, July 11, 2017 at 7:14:23 AM UTC-4, John-Del wrote:   > > thanks   > When projection TVs started hitting the market, manufacturers soon learned that a scan loss in either direction would immediately destroy the CRTs. In response, manufacturers learned to not start the HV until both deflections were established, and shut down same if one was lost or even diminished by a fixed percent. The point is that a vertical collapse would not be the result of a shutdown, but a cause.   > Read my previous post again. One of the last CRT TVs I worked on was a Zenith 35" console with a stubborn vert (slow to fill, weird linearity issues). It turned out that the board became conductive from leaked electrolyte and dust. Heating and cooling the board affected the deflection, but no components were themselves temperature sensitive. I ended up pulling the flyback, smps transformer, and any inductor that looked like it could hold water and immersed the whole board in a parts cleaner. Once assembled, the vertical was perfect and didn't drift.   this board is so clean you could eat off of it. im also looking into that it may be lacking the pulse feed from the flyback to the vert ic.

rickman <gnuarm@gmail.com>: Jul 11 11:38AM -0400 >> Try warming up the board with a hair dryer on high or a heat gun on low and watch the picture as you do. I'd bet dollars to donuts at this point that the picture stretches out. You said you recapped the monitor but was this all the larger electros that are known to cause issues (like a kit) or *everything*. There are often low value electrolytics back at the signal generator side used for signal coupling that get juicy on the bottom and cause issue either from low esr (they can be juicy and still check OK) or from conductive contamination from the leaked electrolyte.   >> If the heat doesn't expose a lazy high esr coupling cap, then try washing the board with a suitable cleaner and acid brush, removing the "jungle" IC if need be but otherwise clean between all the leads.   > well, im starting to think that its going into HV shutdown, and rather than shutting down, the half scan is how it was made to shutdown. still, doesnt shutdown usually not shut down the vertical scan ? i dont think anything on this unit is the usual. the manual for it is near useless, very little info. another thing, the schematic shows two pairs of jumpers, for 13 or 19 inch use. both pairs are jumped on this board, and the same board on ebay has both pairs jumped. its a hair puller. pin 13 on the 7851 is the x ray protection pin, and it needs to be a specific voltage or the chip shuts down. the manual says that, but doesnt say the voltage. im going to end up buying a chasssi, but i'd still like to know what the heck is wrong with mine.   If the high voltage section shuts down you would lose the brightness of the electron beam, no? Sounds from your description the electron beam is clearly visible when the deflection collapses.   --   Rick C

Tom Gardner <spamjunk@blueyonder.co.uk>: Jul 10 07:06PM +0100 On 09/07/17 16:10, Cursitor Doom wrote: > friends describe everything. Check it out:   > https://www.flickr.com/photos/128859641@N02/35684682551/in/dateposted- > public/   It took me <5 minutes to do a spice simulation sufficient to generate such a picture.   The components were an ideal ac voltage source, a diode, an ideal capacitor for smoothing, an ideal resistor for the load, and an ideal resistor in series with the smoothing capacitor representing the smoothing cap's ESR.

Phil Allison <pallison49@gmail.com>: Jul 11 05:30AM -0700 Tom Gardner wrote:   ----------------------- > ideal resistor for the load, and an ideal resistor > in series with the smoothing capacitor representing > the smoothing cap's ESR.   ** The REAL AC supply is not an good sine wave, it is normally somewhat flat topped. Take a look on a scope using a 10:1 probe.   Transformers have significant resistance in their windings, so you need to add some on the AC side of the simulation. The rising part of the saw tooth wave should not be too short, as it will be with zero source impedance.   If you wish to see reality, forget trying to sim it.   In this case grab any unregulated transformer wall wart, put a resistor across the output and scope the voltage across it.   Takes only seconds and leaves no room for doubt.     .... Phil

bitrex <bitrex@de.lete.earthlink.net>: Jul 10 12:50PM -0400 On 07/10/2017 12:21 PM, Cursitor Doom wrote:   >> If you like NIXIs strip them out and use modern driving and build a room >> clock , with a wow factor   > Nice idea that.   Meh, I could buy a couple dozen good Soviet planned-economy strong-like-bear NOS nixies for that price.

Jeff Liebermann <jeffl@cruzio.com>: Jul 10 10:40AM -0700 On Mon, 10 Jul 2017 16:51:34 +0100, Adrian Caspersz   >> <http://www.ebay.com/itm/Vintage-Computer-Measurements-Company-Electronic-Counter-400-MC-Frequency-Conv/271738245865?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649>   >> Asking price is too high either way IMO   >Gut it out. Build in a PC. Sell it to a hipster.   Steampunk conversions are more interesting: <https://www.google.com/search?q=steampunk+computer&tbm=isch>   Take a closer look at red plastic in the panel on the photos. It kinda looks like those are Nixie tubes behind the plastic. That could make it some kind of collectors item. I assume that it does't work since the seller didn't bother plugging it in. So, you could modify it into a large Nixie tube digital clock and power line frequency meter, or just forget about it because it's far too much work.   -- Jeff Liebermann jeffl@cruzio.com 150 Felker St #D http://www.LearnByDestroying.com Santa Cruz CA 95060 http://802.11junk.com Skype: JeffLiebermann AE6KS 831-336-2558

bitrex <bitrex@de.lete.earthlink.net>: Jul 10 04:38PM -0400 On 07/10/2017 01:40 PM, Jeff Liebermann wrote: > since the seller didn't bother plugging it in. So, you could modify > it into a large Nixie tube digital clock and power line frequency > meter, or just forget about it because it's far too much work.   On closer inspection it looks like it's missing a backplane card too. Who knows where one would find a replacement...

You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page. To unsubscribe from this group and stop receiving emails from it send an email to sci.electronics.repair+unsubscribe@googlegroups.com.