Topics

2710 Spectrum Analyser 'Cannot count VCO, IF' error


Martin Whybrow
 

Having repaired the PSU section of my 2710, I'm now trying to deal with the other issues it has. I'm seeing the error message 'Cannot count VCO, IF'; I've checked the counter signals going into the microprocessor board and have measured the following:
J180 - 100MHz at -14dBm, bursts of approx 1us long, repeating every 12us.
J190 - 10 MHz at 0dBm
J195 - 100MHz at -8dBm changing to -4dBm regularly
Can anyone confirm whether or not these signals are what would be expected at the input to the counter, in particular the signal at P180?


Martin Whybrow
 

I've done some probing and have identified a problem; note that mine is a post B010318 model, so it has the CFCS rather than the CFC board.
I have a very low signal level at J130 on the Phaselock Assembly (beat frequency output to the counter); the manual indicates this should be TTL, but it's about 100mV peak to peak. All of the supply lines are in limits, but I'm not seeing any signal gain between J110 (Phase/ Beat signal input) and the output of U720, with ~150mV p-p signal levels at each stage.
I have noted that there are some mods to the board and I'm not sure if thas has any bearing on it, I spotted that C742 and C747 have been removed and a through hole 220nF fitted instead, so this may be a repair to deal with a damaged track.
If anyone's familiar with this assy, I would appreciate your suggestions.


WB6GHK
 

Hi guys, 2710's, oh what fun!

For the "Cannot count VCO, IF" I have found that the majority of the time the 100 MHz timebase module has failed or degraded. The absolute quickest way to determine if it has failed is to sub in a know good working one. It will save you hours of troubleshooting. Find a friend with a later B02xxxx 2710 and beg, steal or borrow the timebase.

Martin, I'm going to assume you have a 2710 service if you troubleshot the power supply, so locate the corresponding pages for the 100MHz timebase and look at the signals coming from the four outputs (I think it's four, going by memory). Scope them to see what is there advise.

I hate to tell you this but B013xxx and below 2710's using the CFCS are not reliable (memory & PS) or accurate (drift), in comparison to later units. Not saying that it isn't worth the time since there is a learning curve and the early 2710 will certainly give you that experience, but the once the unit is operational you may be disappointed.

Best of luck!


Jokken Feldhaar
 

Hi all,

I acquired a 2710 as non-functional, and the 100 MHz Osc had failed. I
put in something cobbled together, and enjoy it working fine.

Cheers, Jochen DH6FAZ

Am 25.11.2020 um 01:13 schrieb WB6GHK:

Hi guys, 2710's, oh what fun!

For the "Cannot count VCO, IF" I have found that the majority of the time the 100 MHz timebase module has failed or degraded. The absolute quickest way to determine if it has failed is to sub in a know good working one. It will save you hours of troubleshooting. Find a friend with a later B02xxxx 2710 and beg, steal or borrow the timebase.

Martin, I'm going to assume you have a 2710 service if you troubleshot the power supply, so locate the corresponding pages for the 100MHz timebase and look at the signals coming from the four outputs (I think it's four, going by memory). Scope them to see what is there advise.

I hate to tell you this but B013xxx and below 2710's using the CFCS are not reliable (memory & PS) or accurate (drift), in comparison to later units. Not saying that it isn't worth the time since there is a learning curve and the early 2710 will certainly give you that experience, but the once the unit is operational you may be disappointed.

Best of luck!




Jaap Rusticus
 

Hello Martin,
Sorry for the late responce. I have an open 2712 here which had repairs and now need adjustments. I checked the signals you asked for:
J195 - signal comes from the Reference Board and was as it should be a stable in amplitude 100MHz signal about 300mV peak-peak, which is about calculated -7dBm.
J190 - 10MHz signal, more or less square wave 450mV peak-peak, which is about 0dBm calculated (including harmonics).
J180 - I measured while "in LOCK" state (in the 2712 lower than in 50kHz/div spans) a constant 800mV peak-peak 16MHz - while not "in LOCK" state there were signal bursts while changing the span/div.
I hope this helps you a bit further. I will have this and other of this family analyzers open for a while to come, so it will be easy for me to do measurements.
Jaap PA0JRK
Netherlands


Jaap Rusticus
 

Hi all,
Nice to meet you.
Just after answering I saw all the other responces. Sorry for not having checked further developments in time.
Jaap


Martin Whybrow
 

Thanks everyone; thankfully mine is a later production version, post B020319, so hopefully not the unrelaible version.
I've checked the 100MHz reference and have healthy outputs on all 3 reference outputs (no tracking generator option installed), roughly -8dBm on J110, -10dBm on J130, and -12dBm on J140 (all approximate as I'm pushing the bandwidth of my cheap scope probe at 100MHz).
One thing I have discovered is that what I thought was an input to the board on J110, the Beat Signal input, was in fact common mode noise from my LED lighting! There appears to be no signal at all coming from the 1st LO buffer, so that's what I'm now looking at.


WB6GHK
 

Martin,
One other thought is compare the cabling to the service manual. I have often seen 271x's that don't work and when someone got into them and found a bigger problem they just closed up the unit and connected the cables to anything that looked close or fit. Lot more than once!


Martin Whybrow
 

The copy of the manual I have is not helpful in regards to cabling, there are no illustrations showing where the various connectors are located. By careful checking of the layout for every board, I was eventually able to determine what goes where and all appeared OK. I did adjust the routing, particularly around the CFC board as a lot of cables cross over each other and I found at least 1 cable was fully snapped into place; that was obviously the problem as it's all working now and appears to be accurate based upon my signal source; I have a higher frequency generator which I'll also try as the one I'm currently using maxs out at 30MHz.
Now I can use it to repair my HP spectrum analyser!


Renée
 

Martin-
I just acquired a 2710...I know it has a battery....Does it store anything important that is lost if battery dies...like those NVRAM in some scope?
Renée

On 11/26/20 1:17 PM, Martin Whybrow wrote:
The copy of the manual I have is not helpful in regards to cabling, there are no illustrations showing where the various connectors are located. By careful checking of the layout for every board, I was eventually able to determine what goes where and all appeared OK. I did adjust the routing, particularly around the CFC board as a lot of cables cross over each other and I found at least 1 cable was fully snapped into place; that was obviously the problem as it's all working now and appears to be accurate based upon my signal source; I have a higher frequency generator which I'll also try as the one I'm currently using maxs out at 30MHz.
Now I can use it to repair my HP spectrum analyser!




WB6GHK
 

Congrats on getting your 2710 working!

In regard to manuals, there are several revisions of the 2710 manual, but try downloading one for the 2712, 2711, 2714, or 2715.

You will find that the analyzer models are very similar in the base design, with certain software or hardware differences depending on the intended market. The quality of the Tektronix manuals varies quite a bit; later ones usually have very good documentation and clarity for schematics and parts layouts, besides narratives explaining the circuit design, function, and operation.


Martin Whybrow
 

Renée, there is a lithium battery on the display & storage board which will probably need replacing; yes, the analyser will lose settings when the battery dies, however a nice feature of these analysers is that they can self calibrate and reset these values; there are some interactive settings, such as display positions that may need setting, but they are well documented in the manual. Let me know if you are having trouble finding a replacement battery, I had to buy several to satisfy the minimum order quantity from Conrad, so I have extras I'd be willing to sell.


WB6GHK
 

Renée,

The lithium battery on the display board and the GPIB/RS232/Centronics board provides backup power for the NVRAM on those boards.

If the display board battery falls below 2.8 VDC (nominally 3.2VDC) it should be replaced. The best way to do this is to use a couple of fresh AA batteries and a holder so you can have a temporary 3 VDC supply source. Remove the board from the analyzer and carefully examine the polarity of the soldered-in battery. Keeping the same polarity as the board battery, solder the AA battery supply leads across the board battery on PCB traces that are away from the board battery so you can unsolder the board battery without disturbing the AA battery supply. Once the board battery is soldered in, remove the AA battery supply and re-install the board in the analyzer. Test the new board battery voltage before and after installing to insure the correct battery voltage.

The idea is to use the AA battery supply to temporarily maintain voltage on the NVRAM while the board battery is being replaced so that calibration settings in the NVRAM are not lost. This way you will not have to go through the service manual procedures to restore accuracy to the analyzer.

NOTE: Even if the lithium board battery is below 2.8 VDC, replace the board battery this way in case the NVRAM possibly still retains the calibration settings.


Renée
 

Thank you.
I was planning on putting 3V across..but was not sure if there was going to be an/ any other issue(s)...you have solved that question. (I had not studied the schematics yet, that is next, service manual does not appear to talk about battery changes....unless I missed it)
currently it is 2.7V..so it needs changed.  i had ordered a tadarian battery with a previous mouser order so I have battery...now to find the time to do the job ASAP.
my apologies for hijacking the thread.....I will go back in the woodwork....
Thank you
Renée

On 11/27/20 10:19 AM, WB6GHK wrote:
Renée,

The lithium battery on the display board and the GPIB/RS232/Centronics board provides backup power for the NVRAM on those boards.

If the display board battery falls below 2.8 VDC (nominally 3.2VDC) it should be replaced. The best way to do this is to use a couple of fresh AA batteries and a holder so you can have a temporary 3 VDC supply source. Remove the board from the analyzer and carefully examine the polarity of the soldered-in battery. Keeping the same polarity as the board battery, solder the AA battery supply leads across the board battery on PCB traces that are away from the board battery so you can unsolder the board battery without disturbing the AA battery supply. Once the board battery is soldered in, remove the AA battery supply and re-install the board in the analyzer. Test the new board battery voltage before and after installing to insure the correct battery voltage.

The idea is to use the AA battery supply to temporarily maintain voltage on the NVRAM while the board battery is being replaced so that calibration settings in the NVRAM are not lost. This way you will not have to go through the service manual procedures to restore accuracy to the analyzer.

NOTE: Even if the lithium board battery is below 2.8 VDC, replace the board battery this way in case the NVRAM possibly still retains the calibration settings.