These have been discussed before at some length. It is not just the output cap that is easily damaged by rough salvage efforts, which however are sadly common enough. There is also a frequently seen problem of bad output spectra with a lot of non harmonic related stuff in the output, not a problem I felt capable of or interested in tracking down. In my sample of two from China via eBay, one was good the other not. Dan

http://www.rbarrios.com/projects/MV89A/ -=Bryan=- > To: time-nuts@febo.com > From: alex@pcscons.com > Date: Tue, 20 Sep 2016 15:09:41 -0700 > Subject: Re: [time-nuts] Normal operating specs of a Morion MV89? > > Hi Brian > > which cap is failing in the MV89 oxco-s? > 73 > > KJ6UHN > > Alex > > On 9/20/2016 2:41 PM, Bryan _ wrote: > > Any tips on resoldering the can, special solder or flux required? Quite often they can be found for a good price on Ebay, from what I have read they are a good OCXO minus the issue of the one particular cap failing. > > > > -=Bryan=- > > > >> From: brayniac@gmail.com > >> Date: Tue, 20 Sep 2016 20:06:59 +0000 > >> To: time-nuts@febo.com > >> Subject: Re: [time-nuts] Normal operating specs of a Morion MV89? > >> > >> Yes. This low output due to capacitor failure is a very common failure. > >> From the last set of MV89a I got on the bay, all had this issue. > >> > >> I had good luck opening mine with a hot air rework station with a small > >> nozzle to blow the solder seal away. A bit of careful prying with a thin > >> flat screwdriver while it is still hot to pop the bottom from the case. The > >> capacitor is easy to get to once the outer can is removed. Hot air to > >> remove. Then tweezers and a fine point soldering pencil made for an easy > >> repair. > >> > >> Obviously, be very careful and be sure to work on an appropriate surface > >> for the amount of heat involved. > >> - Brian > >> > >> On Tue, Sep 20, 2016, 12:57 PM Bryan _ <bpl521@outlook.com> wrote: > >> > >>> May also want to check the output amplitude. If memory serves me correct > >>> it is very common for a capacitor inside this model number to fail. It can > >>> be fixed but requires a torch or a pretty heavy duty soldering iron to get > >>> the can open. > >>> http://www.rbarrios.com/projects/MV89A/ > >>> > >>> -=Bryan=- > >>> > >>>> From: listertim@gmail.com > >>>> Date: Mon, 19 Sep 2016 14:31:06 -0700 > >>>> To: time-nuts@febo.com > >>>> Subject: [time-nuts] Normal operating specs of a Morion MV89? > >>>> > >>>> I recently ordered and received what claims to be a 2007 vintage > >>>> Morion MV89 from ebay (I say claimed as that what the auction item > >>>> said, but my unit is missing the identifying label on the top and just > >>>> has '2007 MV89A' written on the side in marker). I've powered it up > >>>> with a bench power supply at 12V and it's drawing 390mA at steady > >>>> state. I measured the case temperature with a ThermaPen Mk4 (not often > >>>> I get to combine the hobbies of bbq and timenuts...) which is > >>>> calibrated to +/-0.7F to be about 111F, varying somewhat with > >>>> position. > >>>> > >>>> This seems pretty hot - Is this normal ? The datasheet says it should > >>>> draw <350mA when at steady state but doesn't say what temp it should > >>>> be at. Does anyone else have info on their version to see whether this > >>>> is normal ? > >>>> > >>>> Cheers, > >>>> Tim > >>>> _______________________________________________ > >>>> time-nuts mailing list -- time-nuts@febo.com > >>>> To unsubscribe, go to > >>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > >>>> and follow the instructions there. > >>> _______________________________________________ > >>> time-nuts mailing list -- time-nuts@febo.com > >>> To unsubscribe, go to > >>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > >>> and follow the instructions there. > >>> > >> _______________________________________________ > >> time-nuts mailing list -- time-nuts@febo.com > >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > >> and follow the instructions there. > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts@febo.com > > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > > and follow the instructions there. > > > > > > ----- > > No virus found in this message. > > Checked by AVG - www.avg.com > > Version: 2016.0.7797 / Virus Database: 4656/13052 - Release Date: 09/20/16 > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there.

On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: > May also want to check the output amplitude. If memory serves me correct it is very common for a capacitor inside this model number to fail. It can be fixed but requires a torch or a pretty heavy duty soldering iron to get the can open. > http://www.rbarrios.com/projects/MV89A/ > > -=Bryan=- > I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked it up to the output of my MV89A. I get a peak-to-peak measurement of 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if I turn on the FFT mode on the scope I get a peak value of -2.17 db if I am driving the cursor mode correctly. With the 10 MHz from the radioshackus GPSDO used for triggering, it takes ~18 seconds for the MV89 waveform to drift 1 cycle (this is without anything connected on the Uin or Uref pins). Tim

Hi You have one of the many MV89A’s with an output problem. To be absolutely sure of the output, you need to have the scope set to 50 ohm input. If it was set to Hi-Z, the output is likely even lower….. Bob > On Sep 28, 2016, at 1:38 AM, Tim Lister <listertim@gmail.com> wrote: > > On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: >> May also want to check the output amplitude. If memory serves me correct it is very common for a capacitor inside this model number to fail. It can be fixed but requires a torch or a pretty heavy duty soldering iron to get the can open. >> http://www.rbarrios.com/projects/MV89A/ >> >> -=Bryan=- >> > > I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked > it up to the output of my MV89A. I get a peak-to-peak measurement of > 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if > I turn on the FFT mode on the scope I get a peak value of -2.17 db if > I am driving the cursor mode correctly. With the 10 MHz from the > radioshackus GPSDO used for triggering, it takes ~18 seconds for the > MV89 waveform to drift 1 cycle (this is without anything connected on > the Uin or Uref pins). > > Tim > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there.

Take care that the 'scope is not a 50 ohm load: the TDS2024B has not the 50ohm/1Mohm input selection. So far I wasn't able to get from the 'scope FFT the same measurements that I get from a spectrum analyzer. On Wed, Sep 28, 2016 at 7:38 AM, Tim Lister <listertim@gmail.com> wrote: > On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: >> May also want to check the output amplitude. If memory serves me correct it is very common for a capacitor inside this model number to fail. It can be fixed but requires a torch or a pretty heavy duty soldering iron to get the can open. >> http://www.rbarrios.com/projects/MV89A/ >> >> -=Bryan=- >> > > I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked > it up to the output of my MV89A. I get a peak-to-peak measurement of > 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if > I turn on the FFT mode on the scope I get a peak value of -2.17 db if > I am driving the cursor mode correctly. With the 10 MHz from the > radioshackus GPSDO used for triggering, it takes ~18 seconds for the > MV89 waveform to drift 1 cycle (this is without anything connected on > the Uin or Uref pins). > > Tim > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there.

Your oscilloscope fft is plotting single-side amplitude spectrum in V/rtHz (dBV/rtHz), this is in contrast to a spectrum analyzer which plots power spectral density W/Hz or (dBm/Hz). So -2.17 dBv is 10^(-2.17/20) = 0.779 Vrms comparing this to your Vpk-pk measurement 2.3 / 2 / sqrt(2) = 0.813 Vrms That's not to say a scope can't plot power spectrum if the load is explicitly given or assumed 50 Ohms, but in general the plot is amplitude. On Wed, Sep 28, 2016 at 1:38 AM, Tim Lister <listertim@gmail.com> wrote: > On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: > > May also want to check the output amplitude. If memory serves me correct > it is very common for a capacitor inside this model number to fail. It can > be fixed but requires a torch or a pretty heavy duty soldering iron to get > the can open. > > http://www.rbarrios.com/projects/MV89A/ > > > > -=Bryan=- > > > > I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked > it up to the output of my MV89A. I get a peak-to-peak measurement of > 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if > I turn on the FFT mode on the scope I get a peak value of -2.17 db if > I am driving the cursor mode correctly. With the 10 MHz from the > radioshackus GPSDO used for triggering, it takes ~18 seconds for the > MV89 waveform to drift 1 cycle (this is without anything connected on > the Uin or Uref pins). > > Tim > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/ > mailman/listinfo/time-nuts > and follow the instructions there. >

On Wed, Sep 28, 2016 at 4:14 AM, Bob Camp <kb8tq@n1k.org> wrote: > Hi > > You have one of the many MV89A’s with an output problem. To be absolutely > sure of the output, you need to have the scope set to 50 ohm input. If it was set > to Hi-Z, the output is likely even lower….. It seems that the scope is 1 MOhm input impedance and is not changeable. I will have to see if we have any 50 Ohm in-line terminators around that I can also borrow. Does anyone think that there is any value in trying to pursue a return and replacement with the ebay seller (being China, it's likely to be a long round-trip time for both messages and parcels) or just live with it and move on (potentially replacing the capacitor later if I get some way of getting the casing open safely). From what Bob and others have said, it seems that most of the MV89's have this problem so it seems unlikely I would get a better one without a lot of trials. Tim > > Bob > > >> On Sep 28, 2016, at 1:38 AM, Tim Lister <listertim@gmail.com> wrote: >> >> On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: >>> May also want to check the output amplitude. If memory serves me correct it is very common for a capacitor inside this model number to fail. It can be fixed but requires a torch or a pretty heavy duty soldering iron to get the can open. >>> http://www.rbarrios.com/projects/MV89A/ >>> >>> -=Bryan=- >>> >> >> I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked >> it up to the output of my MV89A. I get a peak-to-peak measurement of >> 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if >> I turn on the FFT mode on the scope I get a peak value of -2.17 db if >> I am driving the cursor mode correctly. With the 10 MHz from the >> radioshackus GPSDO used for triggering, it takes ~18 seconds for the >> MV89 waveform to drift 1 cycle (this is without anything connected on >> the Uin or Uref pins). >> >> Tim >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. > > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there.

If you used a 10x probe (10 Meg || 10-15 pF) you may have a pretty weak output, but if you could drive a meter of coax or a 1x probe to 800 mVrms at 10 MHz your probably not too far off spec. 100 pF at 10 MHz is 160 Ohms. On Wed, Sep 28, 2016 at 12:39 PM, Tim Lister <listertim@gmail.com> wrote: > On Wed, Sep 28, 2016 at 4:14 AM, Bob Camp <kb8tq@n1k.org> wrote: > > Hi > > > > You have one of the many MV89A’s with an output problem. To be absolutely > > sure of the output, you need to have the scope set to 50 ohm input. If > it was set > > to Hi-Z, the output is likely even lower….. > > It seems that the scope is 1 MOhm input impedance and is not > changeable. I will have to see if we have any 50 Ohm in-line > terminators around that I can also borrow. Does anyone think that > there is any value in trying to pursue a return and replacement with > the ebay seller (being China, it's likely to be a long round-trip time > for both messages and parcels) or just live with it and move on > (potentially replacing the capacitor later if I get some way of > getting the casing open safely). From what Bob and others have said, > it seems that most of the MV89's have this problem so it seems > unlikely I would get a better one without a lot of trials. > > Tim > > > > > Bob > > > > > >> On Sep 28, 2016, at 1:38 AM, Tim Lister <listertim@gmail.com> wrote: > >> > >> On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: > >>> May also want to check the output amplitude. If memory serves me > correct it is very common for a capacitor inside this model number to fail. > It can be fixed but requires a torch or a pretty heavy duty soldering iron > to get the can open. > >>> http://www.rbarrios.com/projects/MV89A/ > >>> > >>> -=Bryan=- > >>> > >> > >> I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked > >> it up to the output of my MV89A. I get a peak-to-peak measurement of > >> 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if > >> I turn on the FFT mode on the scope I get a peak value of -2.17 db if > >> I am driving the cursor mode correctly. With the 10 MHz from the > >> radioshackus GPSDO used for triggering, it takes ~18 seconds for the > >> MV89 waveform to drift 1 cycle (this is without anything connected on > >> the Uin or Uref pins). > >> > >> Tim > >> _______________________________________________ > >> time-nuts mailing list -- time-nuts@febo.com > >> To unsubscribe, go to https://www.febo.com/cgi-bin/ > mailman/listinfo/time-nuts > >> and follow the instructions there. > > > > _______________________________________________ > > time-nuts mailing list -- time-nuts@febo.com > > To unsubscribe, go to https://www.febo.com/cgi-bin/ > mailman/listinfo/time-nuts > > and follow the instructions there. > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to https://www.febo.com/cgi-bin/ > mailman/listinfo/time-nuts > and follow the instructions there. >

Just put a 47 or 56 ohm resistor from the output pin to ground and look at it with a X10 scope probe. Do you get a sine wave? Over 1 volt pp? Otherwise, you can open up the can and replace the output chip capacitor. Regards ----- Original Message ----- From: "Scott Stobbe" <scott.j.stobbe@gmail.com> To: "Discussion of precise time and frequency measurement" <time-nuts@febo.com> Sent: Wednesday, September 28, 2016 2:38 PM Subject: Re: [time-nuts] Normal operating specs of a Morion MV89? > If you used a 10x probe (10 Meg || 10-15 pF) you may have a pretty weak > output, but if you could drive a meter of coax or a 1x probe to 800 mVrms > at 10 MHz your probably not too far off spec. > > 100 pF at 10 MHz is 160 Ohms. > > On Wed, Sep 28, 2016 at 12:39 PM, Tim Lister <listertim@gmail.com> wrote: > >> On Wed, Sep 28, 2016 at 4:14 AM, Bob Camp <kb8tq@n1k.org> wrote: >> > Hi >> > >> > You have one of the many MV89A’s with an output problem. To be >> > absolutely >> > sure of the output, you need to have the scope set to 50 ohm input. If >> it was set >> > to Hi-Z, the output is likely even lower….. >> >> It seems that the scope is 1 MOhm input impedance and is not >> changeable. I will have to see if we have any 50 Ohm in-line >> terminators around that I can also borrow. Does anyone think that >> there is any value in trying to pursue a return and replacement with >> the ebay seller (being China, it's likely to be a long round-trip time >> for both messages and parcels) or just live with it and move on >> (potentially replacing the capacitor later if I get some way of >> getting the casing open safely). From what Bob and others have said, >> it seems that most of the MV89's have this problem so it seems >> unlikely I would get a better one without a lot of trials. >> >> Tim >> >> > >> > Bob >> > >> > >> >> On Sep 28, 2016, at 1:38 AM, Tim Lister <listertim@gmail.com> wrote: >> >> >> >> On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: >> >>> May also want to check the output amplitude. If memory serves me >> correct it is very common for a capacitor inside this model number to >> fail. >> It can be fixed but requires a torch or a pretty heavy duty soldering >> iron >> to get the can open. >> >>> http://www.rbarrios.com/projects/MV89A/ >> >>> >> >>> -=Bryan=- >> >>> >> >> >> >> I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked >> >> it up to the output of my MV89A. I get a peak-to-peak measurement of >> >> 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if >> >> I turn on the FFT mode on the scope I get a peak value of -2.17 db if >> >> I am driving the cursor mode correctly. With the 10 MHz from the >> >> radioshackus GPSDO used for triggering, it takes ~18 seconds for the >> >> MV89 waveform to drift 1 cycle (this is without anything connected on >> >> the Uin or Uref pins). >> >> >> >> Tim >> >> _______________________________________________ >> >> time-nuts mailing list -- time-nuts@febo.com >> >> To unsubscribe, go to https://www.febo.com/cgi-bin/ >> mailman/listinfo/time-nuts >> >> and follow the instructions there. >> > >> > _______________________________________________ >> > time-nuts mailing list -- time-nuts@febo.com >> > To unsubscribe, go to https://www.febo.com/cgi-bin/ >> mailman/listinfo/time-nuts >> > and follow the instructions there. >> _______________________________________________ >> time-nuts mailing list -- time-nuts@febo.com >> To unsubscribe, go to https://www.febo.com/cgi-bin/ >> mailman/listinfo/time-nuts >> and follow the instructions there. >> > _______________________________________________ > time-nuts mailing list -- time-nuts@febo.com > To unsubscribe, go to > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. >

On Wed, Sep 28, 2016 at 11:46 AM, Tom Miller <tmiller11147@verizon.net> wrote: > Just put a 47 or 56 ohm resistor from the output pin to ground and look at > it with a X10 scope probe. > > Do you get a sine wave? Over 1 volt pp? Hi Tom, Scott, I used about two feet of a (Pomona) coax to minigrabbers test lead and connected it direct to the scope and set the attenuation to 1x in the channel menus. I attached the minigrabbers to the RF and GND pins on the Morion since I didn't fancy holding a probe near the ~50 degree C Morion case.. I got a nice clean sine wave and the in-scope measurement routines reported 2.3 V peak-to-peak. I had 1V/div set so I could fit the OCXO and GPSDO outputs on the scope screen for comparison. (I haven't been able to get a screen capture yet as I need to find a small enough capacity USB key that will be supported by the scope) Tim > > Otherwise, you can open up the can and replace the output chip capacitor. > > > Regards > > ----- Original Message ----- From: "Scott Stobbe" <scott.j.stobbe@gmail.com> > To: "Discussion of precise time and frequency measurement" > <time-nuts@febo.com> > Sent: Wednesday, September 28, 2016 2:38 PM > Subject: Re: [time-nuts] Normal operating specs of a Morion MV89? > > > >> If you used a 10x probe (10 Meg || 10-15 pF) you may have a pretty weak >> output, but if you could drive a meter of coax or a 1x probe to 800 mVrms >> at 10 MHz your probably not too far off spec. >> >> 100 pF at 10 MHz is 160 Ohms. >> >> On Wed, Sep 28, 2016 at 12:39 PM, Tim Lister <listertim@gmail.com> wrote: >> >>> On Wed, Sep 28, 2016 at 4:14 AM, Bob Camp <kb8tq@n1k.org> wrote: >>> > Hi >>> > >>> > You have one of the many MV89A’s with an output problem. To be > >>> > absolutely >>> > sure of the output, you need to have the scope set to 50 ohm input. If >>> it was set >>> > to Hi-Z, the output is likely even lower….. >>> >>> It seems that the scope is 1 MOhm input impedance and is not >>> changeable. I will have to see if we have any 50 Ohm in-line >>> terminators around that I can also borrow. Does anyone think that >>> there is any value in trying to pursue a return and replacement with >>> the ebay seller (being China, it's likely to be a long round-trip time >>> for both messages and parcels) or just live with it and move on >>> (potentially replacing the capacitor later if I get some way of >>> getting the casing open safely). From what Bob and others have said, >>> it seems that most of the MV89's have this problem so it seems >>> unlikely I would get a better one without a lot of trials. >>> >>> Tim >>> >>> > >>> > Bob >>> > >>> > >>> >> On Sep 28, 2016, at 1:38 AM, Tim Lister <listertim@gmail.com> wrote: >>> >> >>> >> On Tue, Sep 20, 2016 at 12:56 PM, Bryan _ <bpl521@outlook.com> wrote: >>> >>> May also want to check the output amplitude. If memory serves me >>> correct it is very common for a capacitor inside this model number to >>> fail. >>> It can be fixed but requires a torch or a pretty heavy duty soldering >>> iron >>> to get the can open. >>> >>> http://www.rbarrios.com/projects/MV89A/ >>> >>> >>> >>> -=Bryan=- >>> >>> >>> >> >>> >> I managed to borrow a Tek TDS 2024B 200 MHz scope from work and hooked >>> >> it up to the output of my MV89A. I get a peak-to-peak measurement of >>> >> 2.3 V which if I have converted it right is ~11 dBm ? Alternatively if >>> >> I turn on the FFT mode on the scope I get a peak value of -2.17 db if >>> >> I am driving the cursor mode correctly. With the 10 MHz from the >>> >> radioshackus GPSDO used for triggering, it takes ~18 seconds for the >>> >> MV89 waveform to drift 1 cycle (this is without anything connected on >>> >> the Uin or Uref pins). >>> >> >>> >> Tim