[time-nuts] Achieving maximum performance when driving 5370A/B inputs

Don Latham djl at montana.com
Sun Feb 28 07:23:15 UTC 2010


Is that buttermilk or blueberry batter? :-)
Don

----- Original Message ----- 
From: "Bruce Griffiths" <bruce.griffiths at xtra.co.nz>
To: "Discussion of precise time and frequency measurement" 
<time-nuts at febo.com>
Sent: Saturday, February 27, 2010 7:32 PM
Subject: Re: [time-nuts] Achieving maximum performance when driving 5370A/B 
inputs


> If one is feeling paranoid about ground loop noise (and wishes to avoid 
> transformers, optoisolators , or fibre optics), etc one could always use 
> an LVDS driver with a batter powered(?) LVDS to CMOS receiver/translator 
> right at the 5370A/B input BNC connector.
> This may be useful for a DMTD system that uses a 5370A/B.
>
> Bruce
>
> Bob Camp wrote:
>> Hi
>>
>> AC cmos will easily drive an L pad to match a 50 ohm cable at these 
>> levels. That's true at either 3.3 or at 5.0 volts. There are a lot of 
>> cmos families out there that beat AC for speed and match the output drive 
>> capability.
>>
>> Bob
>>
>>
>> On Feb 27, 2010, at 9:12 PM, Bruce Griffiths wrote:
>>
>>
>>> 1) One method with 5V CMOS is to add a resistive voltage divider at the 
>>> CMOS driver output with a 50 ohm output impedance at the tap that drives 
>>> the 5370A/B input.
>>>
>>> 2) If one has a 5V 50 ohm driver (eg Thunderbolt PPS output) use a 50 
>>> ohm attenuator at the 5370A/B input.
>>> For a 5370A an attenuation of at least 11dB is required.
>>> For a 5370B an attenuation of at least 3dB is required.
>>>
>>> 3) One can always use the 10x input attenuation setting built in to the 
>>> 5370A/B however this reduces the signal swing to 0.5V at the trigger 
>>> amplifier input (5V CMOS input).
>>>
>>> 4) Attenuate the output of the logic signal by a factor of 2 and use an 
>>> npn emitter follower to drive the 50 ohm load.
>>>
>>> 5) Use 3.3V CMOS signal levels for the 5370B.
>>>
>>> 6) Use a current mode emitter or source coupled switch to drive the 
>>> 5370A/B input.
>>>
>>> The switching jitter of the above drivers will be much lower than the 
>>> internal noise of the 5370A/B as long as HCMOS or faster logic is 
>>> employed.
>>>
>>> Bruce
>>>
>>>
>>> Bob Camp wrote:
>>>
>>>> Hi
>>>>
>>>> Which *still* carefully avoids the issue of how .....
>>>>
>>>> Bob
>>>>
>>>>
>>>> On Feb 27, 2010, at 8:52 PM, Bruce Griffiths wrote:
>>>>
>>>>
>>>>
>>>>> Oops! a small correction (2nd paragraph):
>>>>>
>>>>> For the 5370A attenuating the 5V CMOS signal to a 1V swing with the 
>>>>> threshold set to 0.5V is close to optimum.
>>>>> An input signal with limits of 0V and +1.4V with a trigger threshold 
>>>>> of 0.7V is the maximum usable (for high performance).
>>>>> An input signal with limits of 0V and +0.3V with a trigger threshold 
>>>>> of 0.15V is the minimum usable (for high performance).
>>>>>
>>>>> For the 5370B attenuating the 5V CMOS signal to a 2V swing with the 
>>>>> threshold set to 1V is close to optimum.
>>>>> An input signal with limits of 0V and +3.5V with a trigger threshold 
>>>>> of 0.7V is the maximum usable (for high performance).
>>>>> An input signal with limits of 0V and +0.3V with a trigger threshold 
>>>>> of 0.15V is the minimum usable (for high performance).
>>>>>
>>>>> Thus using the PPS output (~270 ohm is series with a 5V 74AC04 output) 
>>>>> from a Synergy evaluation board that uses an M12M or M12+ GPS timing 
>>>>> receiver to drive the inputs (with a 0-750mV signal) of a 5370A or 
>>>>> 5370B is well within the recommended input signal range for high 
>>>>> performance.
>>>>> This avoids having to adding an external 5V 50 ohm driver that some 
>>>>> would use.
>>>>>
>>>>> Bruce
>>>>>
>>>>> Bob Camp wrote:
>>>>>
>>>>>
>>>>>> Hi
>>>>>>
>>>>>> So exactly how did you know that I bought a (cheap) 5370B a few hours 
>>>>>> ago on the e-place  and was just about to ask about how best to use 
>>>>>> it.
>>>>>>
>>>>>> Hmmmm.......
>>>>>>
>>>>>> Bob
>>>>>>
>>>>>>
>>>>>> On Feb 27, 2010, at 7:01 PM, Bruce Griffiths wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>> The attached excerpts from the 5370A and 5370B manuals indicate that 
>>>>>>> for best performance, that the common practice of driving the 
>>>>>>> 5370A/B 1x inputs directly from a 5V CMOS logic signal is a bad 
>>>>>>> idea.
>>>>>>>
>>>>>>> For the 5370A attenuating the 5V CMOS signal to a 1V swing with the 
>>>>>>> threshold set to 0.5V is close to optimum.
>>>>>>> An input signal with limits of 0V and +1.4V with a trigger threshold 
>>>>>>> of 0.7V is the maximum usable (for high performance).
>>>>>>> An input signal with limits of 0V and +0.3V with a trigger threshold 
>>>>>>> of 0.15V is the minimum usable (for high performance).
>>>>>>>
>>>>>>> For the 5370A attenuating the 5V CMOS signal to a 2V swing with the 
>>>>>>> threshold set to 1V is close to optimum.
>>>>>>> An input signal with limits of 0V and +3.5V with a trigger threshold 
>>>>>>> of 0.7V is the maximum usable (for high performance).
>>>>>>> An input signal with limits of 0V and +0.3V with a trigger threshold 
>>>>>>> of 0.15V is the minimum usable (for high performance).
>>>>>>>
>>>>>>> Bruce
>>>>>>>
>>>>>>> <5370ATriggering.png><5370BTriggering.png>_______________________________________________
>>>>>>> time-nuts mailing list -- time-nuts at 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 at 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 at 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 at 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 at 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 at 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 at febo.com
> To unsubscribe, go to 
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there. 




More information about the time-nuts mailing list