"GRe" wrote in message
l...
"Phil Allison" wrote in message
...
"GRe"
"Phil Allison"
To overcome the issue with fine frequency adjustment, I fitted 500ohm,
10
turn WW pots in series with one of the sections of each Comptom pot -
which is 2 x 15kohms. Makes it easy to set any frequency up to 50kHz,
+/-
1Hz.
Stability is good enough to work with the -100dB notch filter, which at
996 Hz has a notch width of about 0.02 Hz.
Stability issues...
The Leader LAG-120A has a sync terminal for synchronisation with a high
stability frequency source, the circuitry where the synchronisation
takes
place looks very similar to the corresponding part of the BDW141
schematic.
The manual has a description of the sync option (page 7/8):
http://www.slashdocs.com/kwqytp/lag-...on-manual.html
Unfortunatly, other than the original manual I have, there's no
schematic
but with reference to the BDW141 schematic the sync input is simply a
resistor of 10K between the sync terminal and the R4/R5/R54/C14/C17A
node.
Component values are very similar too.
For the LAG-120A in practice the sync range is better than the specified
+/-1%, about +/-2.5% at 1kHz (other freq. not measured).
I don't know about possible pitfalls of implementing such a sync input
for
the BDW but maybe it's worth looking into it.
** Thanks for you interest and effort.
One problem is that feeding synch pulses into the bowels a low distortion
oscillator is BOUND to result in contamination of the sine wave signal.
Probably serious contamination.
The smallest things matter when aiming for a 0.001% THD residual !!
The -100dB notch filter I use for THD measuremenents has somewhat better
stability than the oscillator - since it uses MF resistors and
polystyrene caps for the bridge, plus a couple of WW pots made with
Ni-Chrome wire.
Luckily, one only needs to achieve full null for a few seconds to get a
reliable reading on a scope or millivolt meter.
This is **** simple for THD percentages up to 0.01 % - but becomes a
tad
tedious at 0.001 %.
Until now I have to do with the Leader, with a 0.05%/-66dB THD spec it
was
a better than expected when I measured it last year, 0.014%/-77dB at 1kHz.
So, I can't judge "tediousness" at 0.01% but I assume that's true.
Anyway, stuck with a "bad" oscillator the wish arose for a very low THD
oscillator. It's a project I work on, but with my pace because of all
other
things to do it will take a year.
My best result was circa 0.0005% for my Sony CD101 CD player, playing a
test disk, with a digitally generated sine wave.
Usually I use the LAG120A un-synchronized as with my bench/instruments
setup
it's a PITA to attach the sync cable to the rear panel.
But I remember from some time ago when synchronizing the LAG120A with a
sine
from a HP3325A that distortion was not visible on the o'scope, not even
when
using a square wave as sync signal.
However, that was an "eyeball measurement" and as such does'nt say
anything about possible increasing THD in sync mode.
Probably you're right about increased THD, now I'm curious how well or bad
behaved it is in sync mode.
I'll do a measurement with the HP334A at 1kHz un-synchronized vs.
synchronized and come back with the results.
Here they are.
THD of LAG120A sync & un-sync, range=1kHz x dial=1(!)
Sync signal is 1Vrms/1000Hz/sine/0.05%THD.
0.0149% THD, un-synced
0.0146% THD, synced
The numbers are an average of 4 measurements each, individual readings are
taken from an (averaging) mV-meter connected to the 334A output.
No significant THD difference between synced and un-synced, yet in sync
mode, consistently for all readings, THD is a fraction better.
For the time being I attribute that on absence of short term frequency
instability, any other thoughts?
(!)When experimenting there's always a surprise.
THD of LAG120A @ f=1000Hz BUT range=100Hz x dial=10.
Again sync signal is 1Vrms/1000Hz/sine/0.05%THD.
0.0198% THD, un-synced
0.0189% THD, synced
WTF could be the surprise after unwrapping it?
So, how bad behaved is the generator when provoked?
0.059% THD, synced with ~0.98Vrms/1000Hz/sine/2.4%THD
0.064% THD, synced with 0.1Vpp/1000Hz/square
0.62% THD, synced with 1Vpp/1000Hz/square
Considering the amount of crap injected into the bowels not bad I guess.
Gio Re