EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: Galen on June 05, 2020, 02:35:10 am
-
I tried to connect my TDS3012B to computer via the ethernet port, scope and computer all ON, connect the net cable, then at the scope, Utility->System->I/O->Ethernet Setting, push the DHCP/BOOTP side button to set it from OFF to ON, the scope shows 'Requesting IP address' and a small clock, then it stopped there, no reaction to any button, like dead. I have to disconnect the net cable, then power off the scope.
During the scope is 'Requesting IP address', in the computer network, I can see 'A unknown network connection'. When I pull off the net cable from the scope, the 'Unknown network' on the computer disappeared. So the net cable physically should be OK. I used a good quality cable.
When I restart the scope it is stuck after the splash screen, auto restart again and again. Search the net, found the 'Hard boot' trick, 'Hold trigger B when power on the scope', this worked, scope alive again (language changed back to English).
I'm not a network engineer, so what's the right steps to connect this scope with computer via ethernet port? Especially the configure steps on the computer.
Any advice is very appreciated.
-
I will try to use fixed IP address on the scope, to see if my computer can connect with it.
-
Try fixed IP address, in both computer and the scope, by disable the firewall in computer, still can not communicate. Ping command shows no reply. But the computer local network shows "Connected". Any advice?
-
Is subnetmask and gateway in the same range?
-
If the network cards (of the PC or the instrument (at least one of the two)) are not self-configuring, you may need a "cross' network cable if you directly connect the two devices together, or go through a switch or a router.
This is for hardware side, for software side checks what StefanF rightly said.
-
Thanks a lot StefanF and picburner. The subnet mask is same in both end. I didn't configure gateway. Just link the scope with my computer directly. I think the cable I used is an ordinary one. As picburner advised, I will make a crossover cable, and try again. I think this is the problem. I searched the net, it's said if two host was linked directly, the cable should be crossover cable, pin 1&3,2&6,4&7,5&8 exchanged in the 2 end. I will try this next week.
Thanks again.
-
Tried a crossover cable. still no joy. But another TDS3012B connected by this cable. so the cable is good.
I switched the PSU between the 2 3012B, the no-connection scope still can't connect. Looks like the problem is at the main board?
-
Have you physically checked the eth connector, to see if there is no damage in the pins and or something blocking them?
-
Have you recently repaired/replaced the nvram? (dallas chip)
-
For the NVRAM, yes, I replaced the battery, as the original battery dead. I found the scope power on hour is a very big number. So I think can be considered as 'changed the NVRAM'. Is this related to the network connection?
-
For the connecter scoket, I just try to plug the cable a few times to clean it if any dust/rusty. Not physically check it. normally, it is very rare this thing has problems.
I checked the flexible connection between PSU and the main board, it's good.
-
For the NVRAM, yes, I replaced the battery, as the original battery dead. I found the scope power on hour is a very big number. So I think can be considered as 'changed the NVRAM'. Is this related to the network connection?
In my experience with a 3034b, network will fail if nvram is corrupted.
Mine was working, i removed the nvram to make a backup and replace the battery, and it became corrupted. After that no more network connection!
The mac address shown in the configuration page had changed.
i was able to find a dump somewhere here on the eevblog. There i could see the mac address, i found an entry on my router with the old mac of the scope, put it back in ram, and then it started working. I made tests, and having an invalid mac adress set in ram resulted in no connection, even if the scope with a corrupted mac in nvram, shows a valid mac in the configuration page!
-
There's a button you hold down and power up, I think it's B-trig? That will reset the entire NVRAM to factory defaults, if you suspect it's corrupted that's the first thing I'd try.
-
Hi YetAnotherTechie, I guess your input will work. Because the network working TDS3012B, I didn't do anything to it except changed the crystal oscillator. The other 2 TDS3000B, I de-soldered the NVRAM and replaced the battery. I found one of them the working hours is a very big nunber, So I think the content of it could be 'corrupted'.
So I will try to read out the MAC address from the good one, and write it into the bad one NVRAM using TL866, to test if it works. or I will find a TDS3012B NVRAM image file, in this blog, and write it into the NVRAM. will report back later. thanks a lot for the experience sharing.
-
There's a button you hold down and power up, I think it's B-trig? That will reset the entire NVRAM to factory defaults, if you suspect it's corrupted that's the first thing I'd try.
Hi james, thank you very much for the suggestion. If it's Trigger B, i have tried it before when I try the network connection using DHCP and the scope dead there., have to power off and power on by pushing and holding the trigger B. Otherwise, the scope will be dead after the splash screen. but tried network connection after this, still no joy. Will report any progress.
-
Based on Yetanothertechie's experience sharing, progress was made. I downloaded the 3014B NVRAM image from this blog, and burnt in my 3014B by TL866, the Ethernet connection works well, using fixed IP. I didn't try DHCP mode.
To confirm this, I copyed the NVRAM image from my 3012B, which works for Ethernet, and burn it into the Ethernet problem 3012B, it's Ethernet function recovered.
So the Mac address maybe the cause of the problem.
So I try to only change the Mac address in my old NVRAM file, and burn in my 3012B, Ethernet fail again.
Need to do more study on this. there maybe a verification process in the NVRAM image. Maybe the Checksum?
BTW, the Mac address is stored at x06F0 - x06F5.
Thank you very much Yetanothertechie.
-
If you have 2 NVRAM dumps shouldn't be much difficult to see the checksum. In hindsight, wireshark would have helped...
-
Maybe the MAC address can be set from somewhere? It's possible to replace the NVRAM with a brand new empty one and have the scope work, but I've never tried with a B model so I don't know if a fresh reset puts the MAC address in the NVRAM.
-
Thanks to both YetAnotherTechie and Galen I now have the Ethernet functionality on my TDS3034B restored. :-+
With this occasion, I realized the original contents of the NVRAM had been corrupted and it contained only gibberish.
I used a dump file from somewhere here on the forum and, voila, network connectivity now works like a charm.
-
Thanks to both YetAnotherTechie and Galen I now have the Ethernet functionality on my TDS3034B restored. :-+
With this occasion, I realized the original contents of the NVRAM had been corrupted and it contained only gibberish.
I used a dump file from somewhere here on the forum and, voila, network connectivity now works like a charm.
Looks like I'm in the same loop... my TDS3014B worked fine, until yesterday when I tried to connect it to the network (to hack it to 500MHz, something I've already done multiple times...), I had no luck connecting to the network, although the switch's green led was lighting... tried all possible settings, no luck... and suddenly, the scope wouldn't boot anymore, shortly after the splash screen, it remains frozen, as dead... (cf. picture)
I've tried to push on the 'B Trig' button while powering it on, doesn't help...
I've no clue how to reset the scope's NVRAM... any help/pointers appreciated !!!!
-
I suspect the battery in the Dallas NVRAM has run flat.
And there is no simple way of checking this is actually the case: you need to desolder the NVRAM IC and try to read it with an EEPROM programmer/reader.
-
I suspect the battery in the Dallas NVRAM has run flat.
And there is no simple way of checking this is actually the case: you need to desolder the NVRAM IC and try to read it with an EEPROM programmer/reader.
Don’t think it’s completely flat, because the clock is still running when it’s powered off... there’s some shift, but it still keeps up, more or less...
Will check the datasheet to see if I can read something with a dev board (don’t have any eeprom programmer at hand )
-
Yes, better double check the clock is actually still running, particularly when powered off.
I believe, in my case, the clock would keep time even when powered off, but not advance - e.g. the time would be 19 Sept 2020 15:00 when turning the scope off and, when turning it back on hours later, it would show the same 19 Sept 2020 15:00 or thereabout.
If you need your NVRAM checked, depending on when you are located, you could ship it to me and I could do that for you.
And, if the battery is indeed flat, I could even try to dig it up from the epoxy, solder in a CR2032 battery holder w/ battery and program the NVRAM.
PM me if you are interested.
-
Based on Yetanothertechie's experience sharing, progress was made. I downloaded the 3014B NVRAM image from this blog, and burnt in my 3014B by TL866, the Ethernet connection works well, using fixed IP. I didn't try DHCP mode.
To confirm this, I copyed the NVRAM image from my 3012B, which works for Ethernet, and burn it into the Ethernet problem 3012B, it's Ethernet function recovered.
So the Mac address maybe the cause of the problem.
So I try to only change the Mac address in my old NVRAM file, and burn in my 3012B, Ethernet fail again.
Need to do more study on this. there maybe a verification process in the NVRAM image. Maybe the Checksum?
BTW, the Mac address is stored at x06F0 - x06F5.
Thank you very much Yetanothertechie.
Hi Galen and others,
Thanks to this thread :) I found out what caused loss of Ethernet connection after replacing the battery in my TDS3014B. When I bought the 'scope a few days ago, both RTC and Ethernet connection were flaky. The RTC is fine now.
You said (above) that you used your TL866II to copy the contents of the 3014B file (from elsewhere) into your DS1742W. I have the file but AFAIK, the TL866II does not support the DS1742W directly. Do you remember what target device you specified? Does the DS1742 allow Vcc of 5 V or above? The datasheet suggests it does but I'd rather not kill my DS1742W.
-
Hi Galen and others,
Thanks to this thread :) I found out what caused loss of Ethernet connection after replacing the battery in my TDS3014B. When I bought the 'scope a few days ago, both RTC and Ethernet connection were flaky. The RTC is fine now.
You said (above) that you used your TL866II to copy the contents of the 3014B file (from elsewhere) into your DS1742W. I have the file but AFAIK, the TL866II does not support the DS1742W directly. Do you remember what target device you specified? Does the DS1742 allow Vcc of 5 V or above? The datasheet suggests it does but I'd rather not kill my DS1742W.
Pls select DS1220
-
Hi Galen and others,
Thanks to this thread :) I found out what caused loss of Ethernet connection after replacing the battery in my TDS3014B. When I bought the 'scope a few days ago, both RTC and Ethernet connection were flaky. The RTC is fine now.
You said (above) that you used your TL866II to copy the contents of the 3014B file (from elsewhere) into your DS1742W. I have the file but AFAIK, the TL866II does not support the DS1742W directly. Do you remember what target device you specified? Does the DS1742 allow Vcc of 5 V or above? The datasheet suggests it does but I'd rather not kill my DS1742W.
Pls select DS1220
DS1220 is 5V not 3.3V.
DS1742W is 3.3V
TL866II already has DS1742 in its list but it is also 5V.
-
For what it's worth, I read/wrote a DS1742W with a TL866 before it occurred to me it was giving it 5V. It did no (apparent) damage but had I realized at the time I would not have done this. Later I wired up an adapter to one of my FPGA boards and used the switches and displays on that to manually write a few bytes.
-
For the NVRAM, yes, I replaced the battery, as the original battery dead. I found the scope power on hour is a very big number. So I think can be considered as 'changed the NVRAM'. Is this related to the network connection?
In my experience with a 3034b, network will fail if nvram is corrupted.
Mine was working, i removed the nvram to make a backup and replace the battery, and it became corrupted. After that no more network connection!
The mac address shown in the configuration page had changed.
i was able to find a dump somewhere here on the eevblog. There i could see the mac address, i found an entry on my router with the old mac of the scope, put it back in ram, and then it started working. I made tests, and having an invalid mac adress set in ram resulted in no connection, even if the scope with a corrupted mac in nvram, shows a valid mac in the configuration page!
I think I am having this issue too now, suddenly my 3034 is not reachable over escope (web interface) even though it can print over ethernet just fine. I also niticed something strange, in the IP config (done over DHCP), there is an extra weird character at the end of the domain name, and it comes back even if I delete it manually.
How hard is it to remove the nvram battery and replace it ? do need to take apart the scope entirely ? also I do not seem to be able to find an 3034 nvram dump anywhere here ?
-
see https://www.eevblog.com/forum/testgear/tds-1000-2000-3000-bw-hack/msg4046458/#msg4046458 (https://www.eevblog.com/forum/testgear/tds-1000-2000-3000-bw-hack/msg4046458/#msg4046458) etc.
MAC address in the NVRAM must be 08:00:11:xx:xx:xx
MAC_ADDRESS_IN_NVRAM is 0x06F0.
-
ok so if I take a blank nvram and just program the MAC, that should be enough or do I need an actual original nvram dump ?
I am not finding one for the 3034B so far...
-
Thanks to both YetAnotherTechie and Galen I now have the Ethernet functionality on my TDS3034B restored. :-+
With this occasion, I realized the original contents of the NVRAM had been corrupted and it contained only gibberish.
I used a dump file from somewhere here on the forum and, voila, network connectivity now works like a charm.
I have to do the same for my 3034, which dump file did you use, I can only find an nvram dump for the 3014 ...
are they compatible ?
-
If i remember right then the NVRAM can be reset to factory defaults by pressing B-Trigger on front panel while powering up. The only thing that is not set then is the Ethernet MAC address. That’s because MAC must be unique for each unit.
As mentioned earlier, the MAC is stored in six bytes, and located at DS1728W address 0x6F0. The first three bytes must be 0x08, 0x00, 0x11, the next three you can pick randomly provided the same values are not used in the same network by other equipment.
You’ll need to run all the calibration utilities so that scope inputs DC offsets etc are good again.
-
ok nice to know, but it seems the B-trig on power up does not do anything. I still have my MAC (at least it shows in the menus)
Is the nVRAM format the same for all tds30xx models ?
After I did the "upgrade" to from tds3034 to tds3054, my network is not dead. I mean, it does get a dhcp address, but no printing and no escope anymore. Going to do the DSW conversion...