OCXO/Si5351A Calibration/Testing...Wierd Failure!! #ocxo


Lee Hooper KG5VU
 

Per Ron's suggestion, I have adjusted R21 for startup heater current at about 100 mA.  It is now running WSPR on 30m .... with its case open on the bench and with a separate metered current limited power supply.  The Drift is zero except on the first transmit after restart when I received all 3's!!!!  I plan to leave the U3S on the bench in this configuration for a day or so.

The sensitivity of the adjustment for R21 makes me somewhat nervous.  Per your suggestions, reducing the value of R5 seems smart and will reduce the risk of damaging something.  I will probably do that later in the week.


Stig oz8pz-oz8wsp-oz8wsr
 

Hi, i had the same problem, and it is descibed on qrp-lab Home 
Page when you look for the ocxo. I had adjustet R5 so the Max 
current not too Big.
73, oz8pz, stig 

Sendt fra min iPad

Den 2. maj 2021 kl. 00.36 skrev Lee Hooper KG5VU <mlhooper1@...>:

Based on your initial comments, I set the power supply to the U3S at 5.25V and then used a separate metered power supply for the heater transistors.  This eliminated the problem on the U3S doing repeated resets!!! I was also able to observe the reason for the resets as the current to the heater circuits was extremely high at times.

I played around with this for several hours. At times it appeared to be working correctly, but when I turned of the heater circuit allowing the crystal oven to cool.... then it would either not start to heat again or else it would exhibit the extremely high current phenomena again. 

I have not try restarting the U3S as opposed to just restarting the heater circuits; I can see from your comments that this could be critical.  I am also thinking of reducing the voltage to the heater circuit perhaps by using one or more 1N4001 diodes in series to drop the input voltage.

Can't thank you enough for your time and suggestions!!


Lee Hooper KG5VU
 

Based on your initial comments, I set the power supply to the U3S at 5.25V and then used a separate metered power supply for the heater transistors.  This eliminated the problem on the U3S doing repeated resets!!! I was also able to observe the reason for the resets as the current to the heater circuits was extremely high at times.

I played around with this for several hours. At times it appeared to be working correctly, but when I turned of the heater circuit allowing the crystal oven to cool.... then it would either not start to heat again or else it would exhibit the extremely high current phenomena again. 

I have not try restarting the U3S as opposed to just restarting the heater circuits; I can see from your comments that this could be critical.  I am also thinking of reducing the voltage to the heater circuit perhaps by using one or more 1N4001 diodes in series to drop the input voltage.

Can't thank you enough for your time and suggestions!!


Ron Carr
 

I played around with building and adjusting the OCXO for about 3 weeks.  I was probably overly concerned about drawing too much current and burning out the heater FET's.   I used WSPR drift reports to see how it was working which was a very slow process. 

When turning up R21 by a very small amount,  first there is no change.  Then about 15 seconds later the current will start to rise.  Another 30 seconds and the current will be very high.  Then it will very slowly fall down.

Power off and back on, the current will be very high.   Power off and wait 2 minutes and then back on it will slowly rise.  

In looking at the web page for the kit, in hindsight I think I would try the mod by OZ8PZ, reducing the value of R5.  Also in hindsight if you just adjust for the heater current to be about 100ma, you will be very close to the correct setting.

I also played with shorting C1 where it works as a P controller.   Easy to adjust but I wondered if it would work over the temperature range of my house which would be about 65 in the winter and maybe as high as 95 in the summer.

I also tried using an external resistor to limit the current as is suggested on the web page.  The problem with this approach is that once the heater FET's go into saturation, no heat is generated inside the oven and all the heat is in the external current limiting resistor.


Lee Hooper KG5VU
 

Ron,

Thank you for the comments and suggestions.  

In my unit, I feed 12 VDC to the enclosure which then feeds a home-brew 5VDC supply with a basic LM317 circuit.  I will start by adjusting the pot to raise the output voltage a bit.  If that doesn't do it, I guess that I will test using a second 5VDC supply for the heater circuit.

I can easily imagine that I did advance R21 too aggressively.  How long did it take you to adjust R21 properly.  Hours I suppose???  (I saw another comment on the Group where someone shorted C1 and another that reduced R3 to 100K.  Perhaps one of these can be my backup plan.)

Thanks again,  Lee


Ron Carr
 

I recently built an OCXO kit and encountered similar issues with resets when the unit transmitted.  Using an adjustable supply I found the processor resets at a fairly high voltage.  I don't remember the voltage, but it was higher than 4.75 and higher than I expected.   During experiments the unit did run well on a single supply set at 5.25 volts.  Voltage drop in my power supply wiring may have been an issue and may also be an issue with your unit.

The other thing to keep in mind is that the PI controller has a very long time constant such that when you advance R21 it will take a number of seconds for the current to increase.  If you advance R21 somewhat fast until you see a current increase, you may already have advanced it too far.


Lee Hooper KG5VU
 

I assembled the OCXO kit a few weeks ago.  All seemed to go extremely well until I attempted to set the temperature of the crystal oven.  With the R21 potentiometer fully counterclockwise, the U3S functions as with the standard Si5351A module.  

When I advance the R21 to heat the crystal oven the power supply current begins to pulse rather dramatically at 1-2 second intervals.  The front panel display of the U3S generally goes blank or gives "startup" display. The frequency, monitored on my hf receiver, changes or sometimes the signal disappears completely (Probably the U3S is doing a restart).

I thought this might just be the power supply.  Tried an alternate power supply to no avail.  (I am powering the U3S thru Pin 2 and the heater transistors thru Pin 3 of the 11 pin connector at the bottom of the U3s board but both are connected to a common regulated power supply.). I have not tried using two completely separate power supplies... seems like this should not be necessary.

Has anyone encountered (and hopefully resolved) such a situation?