EMRFD Message Archive 2134
Message Date From Subject 2134 2008-09-26 05:06:38 bkopski AD8307 AT LOW FREQUENCY I have used several AD8307's with great success at rf including in
the QST/EMRFD power meter and SA applications. I'm now attempting to
use it in similar fashion in the several kHz to ~200 kHz area and am
experiencing considerable difficulty and disappointment.
Specifically, the device appears to work as I am used to beginning at
around 200 kHz but begins to deteriorate in performance as I drop the
frequency thru the 10's of kHz and into the high kHz region. Holding
a constant input level but lowering the frequency results in
considerable drop in AD8307 output.
Looking at the output on pin 4, I observe a waveform similar to a
full wave rectified appearance at the higher end of this frequency
range, but as the frequency is lowered this classic waveshape
deteriorates into a very sloppy, unsymmetrical, and unconvincing
waveform shape the "full wave" appearance goes away.
Of course, I've paid attention to all peripheral caps bypasses and
coupling, tried more than one device, and I've floated the input
signal i.e. I lifted the input ground reference from the pc board as
discussed on the AD data sheet no change. Other than the existence
of internal biases on the inputs, I'd normally consider this device
to be operable down to DC, and the AD data sheet even discusses audio
applications. Yet it does not work for me.
Has anyone experience with the AD8307 at these frequencies and / or
any performance issues such as I've experienced?
2136 2008-09-26 12:35:35 Gary Johnson Re: AD8307 AT LOW FREQUENCY See my article in QST, February 2008, p. 33-36, "Accurate S Meter for Direct Conversion
Receivers." You'll see how the 8307 is set up there. One key to LF response is a big cap
from pin 3 to gnd. Fig. 20
2137 2008-09-26 17:59:59 kerrypwr Re: AD8307 AT LOW FREQUENCY I haven't tried to use the 8307 at low frequencies but I've just had a
look at the data sheet.
Fig 38 and the surrounding text relates to audio frequencies; the
figure shows filtering on the output, implying that ripple is high.
This ripple, being related to the input frequency, probably causes
problems when that frequency is low.
Try output filtering.
I home-brewed an audio voltmeter; I used the AD636 which is designed
for audio frequencies.
2138 2008-09-26 18:15:03 wimmie262000 Re: AD8307 AT LOW FREQUENCY > Has anyone experience with the AD8307 at these frequencies and / orHave you placed the two 1uF capacitors as mentioned in the datasheet
> any performance issues such as I've experienced?
> Bob, K3NHI
for LF operation? The offset capacitor on pin 3 is probably needed to
give the expected output waveform at your frequencies. The filter
capacitor on pin 4 you can add after you have trust the operation.
2139 2008-09-27 05:36:55 bkopski Re: AD8307 AT LOW FREQUENCY Hi Joop et al,
My Thanks to you and all the folks here and those who pm'ed me RE: the
low frequency response of the 8307. As so many have pointed out, at
these frequencies a cap is needed on pin 3 and this was an oversight on
my part. Despite soaking myself in the data sheet, I simply missed
this detail. Once I added the cap all was well!
As an incidental, from this pursuit it is clear that no matter the cap
on pin 4, the detected (logged) DC output value is the result of the
input amplitude - period. While a pin 4 cap sets a time constant at
that terminal and of course would affect any waveform appearance there,
it was not relavant to my problem. Cap or no cap - no difference. But
with no cap
2270 2008-10-17 07:17:28 wimmie262000 Re: AD8307 AT LOW FREQUENCY 2272 2008-10-17 09:50:25 bkopski Re: AD8307 AT LOW FREQUENCY 2273 2008-10-17 16:46:30 wimmie262000 Re: AD8307 AT LOW FREQUENCY