[time-nuts] Re: 5370A repair

Tom Verbeure hombre at gmail.com
Fri Jul 18 05:26:43 UTC 2025


Not everything is solved, but I have the 5370A up and running!

* I first disconnected R16 to decouple +15V from the board. After that, I
had an ECL clock on the INT test pin, but no clock to A9 because the board
uses +15V for the ECL to TTL conversion.
* Then I noticed that C17 was indeed burnt. I replaced that with a
different cap and put R16 back in place, but now R16 went up in smoke! So
there's definitely still something wrong on the board.
* My speculation was that the ECL2TTL converter was fine but the 5/10MHz
ext clock to ECL conversion was not. So I removed R16 again, disconnected
R14 (of the ECL2TTL conversion) from its local +15V rail and bodged it to
the incoming 15V on the connector.
* After this, it worked!

The remaining issue now is the ext clock to ECL conversion. This needs to
be fixed if I want to use my external 10MHz reference. There's really not a
lot of components left that can be broken...

I will also replace the Comcor mains input block. I'm convinced now that
its RIFAs are the ones that created the smoke (and the smell.) The 5370A
has more than enough room to put in a replacement that needs more depth to
attach the connections.

Tom


[image: hp5370a_showing_a_result.jpg]

On Thu, Jul 17, 2025 at 11:45 AM paul swed <paulswedb at gmail.com> wrote:

> Tom I have seen the caps short and no smoke. There are 2 15V caps C17
> before R16 and C16 after. Either could cause the issue. Simply lift a lead
> on both caps and see if the system works.
> Paul
> WB8TSL
>
> On Thu, Jul 17, 2025 at 1:36 PM Tom Verbeure <hombre at gmail.com> wrote:
>
>>
>>
>> On Thu, Jul 17, 2025 at 10:00 AM paul swed <paulswedb at gmail.com> wrote:
>>
>>> Tom check to see if the 10 MHz feeds the 100 MHz multiplier. I think
>>> that ultimately drives the system.
>>> Without the board I think the clocks are missing.
>>>
>>
>> Of course! I should have thought about that.
>>
>> Right now, I don't even have anything on the "INT" or on the "EXT" test
>> points on the A8 reference frequency buffer board. At least that narrows
>> the issue down to this board only for now.
>>
>> I don't see anything obvious on the board where a failure would result in
>> shorting the +15V, but one useful thing is that the +15V is only used when
>> there's an EXT clock in play. So I could disconnect the +15V (e.g. by
>> removing R16 in the ripple filter) and then check if the INT clock comes
>> through.
>>
>> Thanks!
>> Tom
>>
>>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: hp5370a_showing_a_result.jpg
Type: image/jpeg
Size: 254476 bytes
Desc: not available
URL: <http://febo.com/pipermail/time-nuts_lists.febo.com/attachments/20250717/2490c45e/attachment.jpg>


More information about the Time-nuts_lists.febo.com mailing list