[time-nuts] Opening an Isotemp OCXO
preilley_454 at comcast.net
Fri Oct 28 15:05:27 EDT 2016
There is no regulator in the unit. The power pin is connected directly
to the S30 chip.
On 10/28/2016 1:31 PM, J. L. Trantham wrote:
> Have you measured the voltage on the 'power' pin for the chip with 12 V applied to the OCXO (or 5 V applied)?
> Is there a 5 V regulator on the board?
> -----Original Message-----
> From: Peter Reilley [mailto:preilley_454 at comcast.net]
> Sent: Friday, October 28, 2016 9:11 AM
> To: J. L. Trantham
> Subject: Re: [time-nuts] Opening an Isotemp OCXO
> Thanks for the link. I did not find any S30 chips that would run off
> 12 volts.
> Could the whole OCXO be a 5 volt unit?
> On 10/28/2016 9:53 AM, J. L. Trantham wrote:
>> Thanks for the update.
>> No time to spend right now but I found this by googling 'TI S30 SOIC chip'
>> Hope it helps.
>> -----Original Message-----
>> From: Peter Reilley [mailto:preilley_454 at comcast.net]
>> Sent: Friday, October 28, 2016 8:24 AM
>> To: J. L. Trantham
>> Subject: Re: [time-nuts] Opening an Isotemp OCXO
>> I did finally get it open. I used a very large old style soldering
>> iron and .003 inch steel
>> shim stock. I would melt the solder on the straight seams and insert
>> small pieces of
>> the shim. Solder does not stick well to steel so the shim kept the
>> soldered seam open.
>> I used a soldering iron rather than a torch because I can control the temperature.
>> I could not use the shim at the corners. After all the straight seams
>> were separated
>> I could pull each corner using a screw in the mounting hole and melt the solder at
>> the corner. Slowly working my way around, corner by corner, I got it
>> opened. I did
>> not damage anything so I should be able to close it up after I fix it.
>> Looking around with my scope it seems that the output driver chip is bad as I expected.
>> It is a TI 14 pin surface mount DIP. It says S30 on it which if it is
>> a 74S30 it is an
>> 8 input positive NAND gate. The board layout confirms this as the 10
>> MHz signal is
>> connected to pin 2 and all other inputs are tied high. Pin 8 is
>> connected to the output.
>> The chip is run off 12 volts so it must be CMOS. But I cannot find any
>> chip like that
>> that will run off 12 volts. Any suggestions for a replacement?
>> Also, using an 8 input NAND chip for a driver seems an odd choice.
>> When I put 12 volts on the unit the S30 chip gets really hot. After I removed the chip
>> the unit seems to work OK. The current jumps between about .1 amp to
>> .9 amps. It seems
>> like the temperature regulator is an on/off type controller.
>> The device on eBay, item 261920574725, looks exactly like what I have.
>> I have placed a bunch of pictures in my dropbox.
>> On 10/18/2016 10:57 AM, J. L. Trantham wrote:
>>> I'm not familiar with your OCXO but I found one shown on 'theBay' (item 261920574725) and it appeared to have an option for 'mounting screws', four of them, on the bottom. Interestingly, the 'link' to the datasheet for that unit did not show threads for mounting screws.
>>> If your unit has that option, I would suggest placing four long screws, mounting the item in a vise, use a small torch (I've used a hand held propane torch turned down very low to open a number of units from 5061A's) around the bottom of the case while gripping the top with an appropriate sized Channel Lock plier and lifting off the top.
>>> If you can repair the OCXO, it should be easy to reassemble the unit with solder.
>>> TheBay unit looks like it has a screw cover (which likely has a rubber gasket) for mechanical adjustment of the frequency. I'd remove that before applying the torch. :^).
>>> If you get it open, I'd love to see some pictures of the insides.
>>> Good luck and hope this helps.
>>> -----Original Message-----
>>> From: time-nuts [mailto:time-nuts-bounces at febo.com] On Behalf Of
>>> Peter Reilley
>>> Sent: Tuesday, October 18, 2016 8:11 AM
>>> To: time-nuts at febo.com
>>> Subject: [time-nuts] Opening an Isotemp OCXO
>>> I bought an Isotemp OCXO82-59 with a frequency of 10 MHz for a $3 at the MIT flea market.
>>> As expected it was dead. It heats up as expected but looking at the
>>> output with a scope there
>>> is nothing. However looking at the output with a spectrum analyzer I
>>> can see a faint 10 MHz
>>> signal. It seems that the oscillator is running but the output
>>> circuitry is dead. Reasonable
>>> Anyway, has anyone had any luck unsoldering the tin case without destroying it?
>>> time-nuts mailing list -- time-nuts at febo.com To unsubscribe, go to
>>> and follow the instructions there.
More information about the time-nuts