[time-nuts] Advantages & Disadvantages of the TPLL Method

[WarrenS]

Steve posted

> Add smilies as you see fit.
> But can it make a decent cup of char?
> Can you add a cup-holder, those are essential these days.
> So does it grind wheat to make flower then; by golly, this thing is 
> versatile.

Funny or sarcastic, depending where one puts the smilies, but I would not 
yet count anything out. :-)
A major limitation for me with the simple TPLL has been it only test things 
that are very close in freq to it's reference Frequency, which I mistakenly 
considered was the same as it's controlled Ref oscillator .

Necessity can be the mother of innovation
I recently had a need to test an off freq Osc that was not within range of 
the 10811 Osc I normally use with my TPLL.
Simple solution turned out to be.
Use the TPLL's  controlled 10811 as the reference for my freq synthesizer 
and use the fully adjustable synthesizer output as the reference freq of the 
TPLL.  After a couple of minor changes in level and Bandwidth, I now have a 
full frequency range TPLL tester.
OF course the performance is still limited by the quality and B/W of the 
TPLL's reference signal which now must include any synthesizer limitations, 
BUT being limited to a single fixed frequency is no longer one of the 
limitations.

IN fact it is now obvious to me that ANY reference freq can be used with the 
simple TPLL, as long as it is derived in some way from a EFC controlled OSC. 
The controlled oscillator freq can be divided, multiplied, mixed up or down 
or used with a synthesizer, same as any PLL, and this opens up a whole new 
range of uses for the TPLL. (and a new set of trade offs)
No longer a $10 solution, but then a had an eBay synthesizer setting around 
already, so it did not add any extra cost for me.

ws

******************
********************

[time-nuts] Advantages & Disadvantages of the TPLL Method
Steve Rooke sar10538 at gmail.com
Mon Jun 14 02:37:32 UTC 2010

Warren,

Penny dropped!

On 14 June 2010 05:29, WarrenS <warrensjmail-one at yahoo.com> wrote:
> Magnus Posted
>
>> "The TPLL is a mid-tau stability test, since it's sweet-spot is in the
>> 0,1 - 1000 s range. Short-term is better handled in LPLL phase-noise
>> measurements."
>
> Mostly agree, it is a mid-tau device, but also consider:
>
> The simple TPLL can be used to find the 1ms to 10 ms ADEV as good and 
> maybe?
> better than ANYthing else.
> AND I use it to find the drift and ADEV of my dual oven HP 10811's out to
> weeks, so there is no long term limit.
> (the trick is to control the DUT and not the Reference when doing long 
> term
> testing)

Of course, provided that the DUT has an EFC, you can put this in place
of the "reference" oscillator and put a, say, GPSDO in place of where
the DUT normally goes. This will certainly extend things at long Tau
out as far as you wish. As for short tau, it would be better to
convert the circuit to LPLL.

> BUT
> I find the TPLL method is useful for MUCH more than just getting ADEV 
> data.
> As such it's "sweet spot" extends from below 0.004 ms  (250 Hz)
> all the way up to >> weeks, when the DUT is the controlled Osc.

Either 0.004 s or 4 ms methinks. You would have to make sure that the
B/W from the mixer out right to the ability to control the reference
oscillator is that wide. This could be verified by breaking off the
EFC and injecting 250Hz into there and seeing if the original feed to
the EFC follows it.

> ALSO the simple TPLL is very Good and useful for seeing and measuring the
> effects of temperature, line noise, vibration, PS,  load, and just about
> everything else that effects the frequency stability of an oscillators 
> below
> 1 KHz or so.

Providing that the reference oscillator is not subjected to those
effects at the same time (isolation).

> So if you are just thinking about the TPLL for taking ADEV data from 0.1 
> to
> 1000 sec, then you're are missing 90% of the other useful stuff it can do 
> as
> good or better than most anything thing else out there, and all for the 
> same
> $10 (my cost).

If you have some "stock" items that you can borrow for this job and
not just dedicate it to the TPLL. Still, that is not an unreasonable
assumption for many.

> If still in doubt about some of the TPLL's other high end performance
> capabilities, try the swinging osc test with ANYthing else and see if you
> can match the simple TPLL's performance.
>
> If one would make a list for the non ADEV uses and advantages of the 
> simple
> TPLL
> It would indeed be very long list, and it does not take any fancy write up
> to see that, just a bit of open minded thinking.

But can it make a decent cup of char?

> BTW, with a couple of minor configuration changes, the TPLL BreadBoard can
> be transformed into a LPLL,
> so the usefulness of the basic Universal TPLL BB circuit has even more
> possibilities.
> In fact one could make yet another list of all the additional things it 
> can
> do with no added cost,

Can you add a cup-holder, those are essential these days.

> just by changing a few jumpers and values.
> But those things are for later discussions, one windmill at a time.

So does it grind wheat to make flower then; by golly, this thing is 
versatile.

Add smilies as you see fit.

Cheers,
Steve

> ws
>
>