Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Post by Richar » Fri, 28 Mar 2003 21:02:30

Hi. Just wondering if anyone out there has changed the original
Drake RX 1st IF filter and therefore has one available.  Am
musing over the idea of whether to use this filter in my RX.

If you have the a filter perhaps you could just drop me a post,
then I can know if I can get hold of this filter.  Please
indicate what you might be wanting for it (hopefully cheap)  :c).

My email:


Please remove "REMOVE".

Anyone have the specs for this filter?  I know it's about 8 Khz
bandwidth.  I don't know terminating impedances.

TIA. Rich.

 
 
 

Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Post by Floyd Davidso » Sat, 29 Mar 2003 00:18:22



>Hi. Just wondering if anyone out there has changed the original
>Drake RX 1st IF filter and therefore has one available.  Am
>musing over the idea of whether to use this filter in my RX.

That filter is generally thought to be the biggest impediment
to the stock R4C being a good receiver.  I would assume it was
a marketing decision, intended to make it more attractive by
lowering the price.

If you are not going to market your receiver, you probably don't
want to use that particular filter!

Instead, either watch eBay for one of the replacement filters,
or you might look into some of the AM filters for other receivers.

A quick search in eBay came up with current examples:  

  6 kHz / 5.645 mHz for Drake R4C
  6 kHz / 9.010 mHz for ICOM
  6 kHz / 3.180 mHz for Yaesu FT-101

A search of past auctions gives an idea of prices:

  6 kHz / 5.645 mHz for Drake TR-7       $ 90.00
  6 kHz / 5.645 mHz for Drake TR-7         87.00
  6 kHz / 5.645 mHz for Drake TR-7         84.99
  6 kHz / 5.645 mHz for Drake R4C         111.05
  6 kHz / 5.645 mHz for Drake R4C         102.50
  6 kHz / 8.830 mHz for Kenwood TS-430S    62.00
  6 kHz / 8.830 mHz for Kenwood TS-430S    65.00

A google search came up with other examples:

  8 kHz / 5.645 mHz for Drake TR-7

Plus of course there are the commonly available 6 kHz and 8 kHz
replacement filters for the R4C available for about $105 each
from INRAD.

And you might want to consider more than one filter and
switching between them.  Both the 8 kHz and 6 kHz filters are
wider than necessary for SSB or CW and using switchable narrower
filters is probably a *much* preferable system from a
performance point of view.  The only benefits to using the
wide filter early in the IF chain are 1) lower cost compared to
switching filters, 2) transient response for the Noise Blanker,
and 3) passband tuning of the entire second filter bandwidth.
If you don't need 2 or 3, the cost of switching between narrower
filters is probably worth the performance gain for a one time
only receiver project.

If you go with narrow filters, Ten-Tec uses filters at 9001.5
and 6298.5 kHz, Heathkit used HF filters, and I suppose there
are many many others..  INRAD, for example, sells a 2.8 kHz
filters that could be followed by almost any of the common
narrower (2.4, 2.3, 2.1, 1.8 etc.) filters.  For CW a 600 Hz
filter can be used in front of a 600 Hz filter, or use 500/400,
400/300, or 300/250 combinations depending on what you can find.

Most of the INRAD filters have information available on them
at <http://www.qth.com/INRAD/>, and they list the frequencies
of various replacement filters which can be used sometimes to
identify less than well described used filters found for sale
in other places.

Quote:>If you have the a filter perhaps you could just drop me a post,
>then I can know if I can get hold of this filter.  Please
>indicate what you might be wanting for it (hopefully cheap)  :c).

With filters though, you are going to get what you pay for.  If
it will do you well enough, INRAD has $24 filters too. :-)

   <http://www.qth.com/INRAD/exp_filt.htm>

--
Floyd L. Davidson         <http://www.ptialaska.net/~floyd>

 
 
 

Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Post by Floyd Davidso » Sat, 29 Mar 2003 07:31:53


>I spent about $160 for the Yaesu 50 line, that is FL50B, FR50B,
>FV50B.  It is a budget system really, but quite good.  A
>beginners set. However, I'm trying to spend not very much at all
>improving 2nd mixer cross mod as well as improving the main
>selectivity (acheived at 455 Khz, 2nd IF). Heck I only paid $160
>for the entire set-up! :c)

OK, now I've got a better idea what you are doing!

I did a web search looking for a manual for the receiver, and
didn't find anything.  Do you know of an online description or
manual that gives details on the IF and mixer schemes?

Quote:>I've thought the filter to help stop the 2nd mixer from cross
>modulation needs to be 8 Khz wide. Such a filter will pass
>SSB/CW/AM, the real bandpass shaping takng place at the 2nd IF.
>But, actually, maybe I could go with a 6 Khz filter.  For some
>reason the filter to "protect" mixer cross mod in the Drake RX's
>is 8khz wide, and I've taken my cue from this fact.

In the Drake design the 1st IF filter has to be at least twice
as wide as the widest 2nd IF filter because there is only one
1st IF filter and it has to pass _both_ sidebands, only one of
which is selected by the 2nd IF filter, but since either of them
could be, they both have to be there.  The actual width has to
be a little wider than that because, or at least I think it is
because, the noise blanker's gate is after the filter instead of
before it.

Hence, you don't necessarily have the same design criteria that
Drake did (though it is entirely possible that in fact you will
end up with the same requirements).

Quote:>I think using a wide filter in the 1st IF would allow for
>passband tuning as you mention below. And I was thinking about
>arranging for passband tuning. I was hoping to have a good SSB
>and a CW filter in the 2nd IF.

But you also (as is obvious from the commonly derogatory
comments about the original Drake filter) want to have
selectivity as early as possible.  The difference may or may not
affect you though.  For example, sitting here at the top of the
world I don't exactly have to pick signals out of a crowded
band.  I hear more aur***buzz than I do QRM from the one other
ham in town that operates HF.  Maybe the tighter specs on a
roofing filter rarely do me any good at all! :-)

But for someone near an urban center (particularly if they are
"smart" and found a great location at the highest elevation
around) that extra 20 to 40 dB of floor on a good filter will be
the difference between ability to get into DX pileups or operate
contests, or not.

Quote:>The Drake filter would improve things, maybe I could get one
>really cheap, but would it do the job.  That I don't know.  Bit
>of a risk maybe, but if I could get one cheap maybe it's worth
>the risk.

One advantage to doing it that way is the low expense of testing
your theory...  and if it works but needs a better filter _then_
you can spend the money on a drop in replacement.  Anything else
you do won't be a drop in replacement so much as a total (ouch)
redesign...

Quote:>Might actually be best making my own filter, but will look at
>"store bought" for a while.

It's one of those "variables", that goes into the entire juggling
act. :-)  There are tradeoffs...

Quote:>Thanks for the info on prices. I'll keep searching Ebay and see
>what comes up and for what price it sells for.

Hmmm...  I don't see very many "cheap" filters on eBay, or
anywhere else.  They all seem to go for premium rates.  Buying
two filters would more than double your investment!  But that
might be just the way to go.

Here are some thoughts:

If you don't need AM, you don't need anything wider than just
more than twice your SSB filter bandwidth.

If you don't use the same passband tuning design that Drake did,
you probably don't need a 1st IF filter to handle two sidebands.
I don't find the passband tuning to be very useful for SSB, but
for RTTY and CW it is essential.  Which suggests that perhaps a
3.0 kHz 1st IF filter would be just fine, even though it would
provide only slight adjustment of the passband on SSB.  (Hmmm...
maybe I should try that myself!)

Also, you can set up a passband tuning system that uses only two
filters one at the 1st IF and one in the 2nd IF, that are
offsetting as the passband is tuned, thus giving you variable
bandwidth with only two filters.  I previously posted a
description of the most versatile way that could be done, but it
could also be done using only the passband tuning to also
control the offset between the two filters and thus the
bandwidth. (It could also be done using only 2 IF's but then the
"center frequency" changes as things get tuned.)

I take it you are going to replace all of the oscillators and
mixers, hence the entire frequency scheme is up for grabs as
part of this project?  (Find either a Collins PTO or a Drake
PTO, you'll be lots happier with the stability.)  I really do
like the basic scheme that Drake used, but it requires some very
careful attention to detail for the AGC too.  In particular the
RF and 1st mixer need to have good dynamic range and appropriate
AGC to keep the signal levels under control before they hit the
later mixer stages.  (That's why they used a remote cutoff tube,
the 6BA6, for the RF stage instead of a sharp cutoff that is so
commonly seen.)

--
Floyd L. Davidson         <http://www.redwaveradio.com/~floyd>

 
 
 

Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Post by Richar » Sat, 29 Mar 2003 08:37:31



> >I spent about $160 for the Yaesu 50 line, that is FL50B,
FR50B,
> >FV50B.  It is a budget system really, but quite good.  A
> >beginners set. However, I'm trying to spend not very much at
all
> >improving 2nd mixer cross mod as well as improving the main
> >selectivity (acheived at 455 Khz, 2nd IF). Heck I only paid
$160
> >for the entire set-up! :c)

> OK, now I've got a better idea what you are doing!

> I did a web search looking for a manual for the receiver, and
> didn't find anything.  Do you know of an online description or
> manual that gives details on the IF and mixer schemes?

I have the schematic on my PC.

- Show quoted text -

Quote:> >I've thought the filter to help stop the 2nd mixer from cross
> >modulation needs to be 8 Khz wide. Such a filter will pass
> >SSB/CW/AM, the real bandpass shaping takng place at the 2nd
IF.
> >But, actually, maybe I could go with a 6 Khz filter.  For some
> >reason the filter to "protect" mixer cross mod in the Drake
RX's
> >is 8khz wide, and I've taken my cue from this fact.

> In the Drake design the 1st IF filter has to be at least twice
> as wide as the widest 2nd IF filter because there is only one
> 1st IF filter and it has to pass _both_ sidebands, only one of
> which is selected by the 2nd IF filter, but since either of
them
> could be, they both have to be there.  The actual width has to
> be a little wider than that because, or at least I think it is
> because, the noise blanker's gate is after the filter instead
of
> before it.

Actually, with the FR50B, if you only want to receive LSB on
80/40, and USB on 20/15/10 that 1st IF filter can be less than 8
Khz.  You only need 8 Khz if on any particular band you want to
switch sidebands and you don't alter the VFO frequency.  (You
don't really want to alter the VFO if a station went from USB to
LSB).  So, I'd say a 6Khz filter would be okay.  Because of
frequency translation being different for 80/40 and for 20/15/10,
LSB (8/40) and USB (20.15/10) sidebands always land within the 3
Khz filter passband at 455Khz.  Switching sidebands on a
particular band would call for two SSB filiters at 455 Khz, or
alteration of the 2nd mixer xtal osc frequency.  I intend to
alter the 2nd mixer xtal osc to cater for CW.  To put the CW
carrier on 455 Khz, rather than just off the centre frequency as
with SSB.

- Show quoted text -

Quote:> Hence, you don't necessarily have the same design criteria that
> Drake did (though it is entirely possible that in fact you will
> end up with the same requirements).

> >I think using a wide filter in the 1st IF would allow for
> >passband tuning as you mention below. And I was thinking about
> >arranging for passband tuning. I was hoping to have a good SSB
> >and a CW filter in the 2nd IF.

> But you also (as is obvious from the commonly derogatory
> comments about the original Drake filter) want to have
> selectivity as early as possible.  The difference may or may
not
> affect you though.  For example, sitting here at the top of the
> world I don't exactly have to pick signals out of a crowded
> band.  I hear more aur***buzz than I do QRM from the one
other
> ham in town that operates HF.  Maybe the tighter specs on a
> roofing filter rarely do me any good at all! :-)

> But for someone near an urban center (particularly if they are
> "smart" and found a great location at the highest elevation
> around) that extra 20 to 40 dB of floor on a good filter will
be
> the difference between ability to get into DX pileups or
operate
> contests, or not.

> >The Drake filter would improve things, maybe I could get one
> >really cheap, but would it do the job.  That I don't know.
Bit
> >of a risk maybe, but if I could get one cheap maybe it's worth
> >the risk.

> One advantage to doing it that way is the low expense of
testing
> your theory...  and if it works but needs a better filter
_then_
> you can spend the money on a drop in replacement.  Anything
else
> you do won't be a drop in replacement so much as a total (ouch)
> redesign...

> >Might actually be best making my own filter, but will look at
> >"store bought" for a while.

> It's one of those "variables", that goes into the entire
juggling
> act. :-)  There are tradeoffs...

> >Thanks for the info on prices. I'll keep searching Ebay and
see
> >what comes up and for what price it sells for.

> Hmmm...  I don't see very many "cheap" filters on eBay, or
> anywhere else.  They all seem to go for premium rates.  Buying
> two filters would more than double your investment!  But that
> might be just the way to go.

> Here are some thoughts:

> If you don't need AM, you don't need anything wider than just
> more than twice your SSB filter bandwidth.

Or not even that if you stick to LSB on 80/40, USB on 20/15/10.

Quote:

> If you don't use the same passband tuning design that Drake
did,
> you probably don't need a 1st IF filter to handle two
sidebands.
> I don't find the passband tuning to be very useful for SSB, but
> for RTTY and CW it is essential.  Which suggests that perhaps a
> 3.0 kHz 1st IF filter would be just fine, even though it would
> provide only slight adjustment of the passband on SSB.
(Hmmm...
> maybe I should try that myself!)

I bet I could get away with say a 4 Khz filter.

- Show quoted text -

Quote:> Also, you can set up a passband tuning system that uses only
two
> filters one at the 1st IF and one in the 2nd IF, that are
> offsetting as the passband is tuned, thus giving you variable
> bandwidth with only two filters.  I previously posted a
> description of the most versatile way that could be done, but
it
> could also be done using only the passband tuning to also
> control the offset between the two filters and thus the
> bandwidth. (It could also be done using only 2 IF's but then
the
> "center frequency" changes as things get tuned.)

> I take it you are going to replace all of the oscillators and
> mixers, hence the entire frequency scheme is up for grabs as
> part of this project?  (Find either a Collins PTO or a Drake
> PTO, you'll be lots happier with the stability.)  I really do
> like the basic scheme that Drake used, but it requires some
very
> careful attention to detail for the AGC too.  In particular the
> RF and 1st mixer need to have good dynamic range and
appropriate
> AGC to keep the signal levels under control before they hit the
> later mixer stages.  (That's why they used a remote cutoff
tube,
> the 6BA6, for the RF stage instead of a sharp cutoff that is so
> commonly seen.)

What I was thinking is this:

Assuming that VFO is not changed when transmiting station goes
from SSB to CW:

Have a 1st IF with BW say anything from 4 to 8 Khz.  Wider
bandwidth allows for greater passband tuning.  (Probably only
need about 5 Khz with the FR50B if not wanting to change
sidebands on a particlar band).

Use a 2.5 Khz SSB filter centered on 455 Khz and CW filter
centered on 455 Khz.  Have 2nd mixer be a VXO with varicap that
can be switched in, or with potentiometer marked SSB and CW. When
on SSB the carrier lands off-center of 455 Khz, (say 456.5).
When on CW flick switch or adjust pot. and that puts carrier on
455 Khz dead center of the CW filter.  Vary pot. and that will
have the effect of moving the SSB signal across passband of SSB
filter.

I think the above is passband tuning.

 
 
 

Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Post by Richar » Sat, 29 Mar 2003 08:44:08


Quote:

> What I was thinking is this:

> Assuming that VFO is not changed when transmiting station goes
> from SSB to CW:

> Have a 1st IF with BW say anything from 4 to 8 Khz.  Wider
> bandwidth allows for greater passband tuning.  (Probably only
> need about 5 Khz with the FR50B if not wanting to change
> sidebands on a particlar band).

> Use a 2.5 Khz SSB filter centered on 455 Khz and CW filter
> centered on 455 Khz.  Have 2nd mixer be a VXO with varicap that
> can be switched in, or with potentiometer marked SSB and CW.
When
> on SSB the carrier lands off-center of 455 Khz, (say 456.5).
> When on CW flick switch or adjust pot. and that puts carrier on
> 455 Khz dead center of the CW filter.  Vary pot. and that will
> have the effect of moving the SSB signal across passband of SSB
> filter.

> I think the above is passband tuning.

But, the BFO signal must also track the sideband.  I must figure
that out.
 
 
 

Testing the waters: DRAKE 5645 Khz 4 pole xtal filter BW 8 Khz.

Post by Richar » Sat, 29 Mar 2003 09:01:38




> > What I was thinking is this:

> > Assuming that VFO is not changed when transmiting station
goes
> > from SSB to CW:

> > Have a 1st IF with BW say anything from 4 to 8 Khz.  Wider
> > bandwidth allows for greater passband tuning.  (Probably only
> > need about 5 Khz with the FR50B if not wanting to change
> > sidebands on a particlar band).

> > Use a 2.5 Khz SSB filter centered on 455 Khz and CW filter
> > centered on 455 Khz.  Have 2nd mixer be a VXO with varicap
that
> > can be switched in, or with potentiometer marked SSB and CW.
> When
> > on SSB the carrier lands off-center of 455 Khz, (say 456.5).
> > When on CW flick switch or adjust pot. and that puts carrier
on
> > 455 Khz dead center of the CW filter.  Vary pot. and that
will
> > have the effect of moving the SSB signal across passband of
SSB
> > filter.

> > I think the above is passband tuning.

> But, the BFO signal must also track the sideband.  I must
figure
> that out.

SSB

You have 2nd mixer VXO set to SSB. You have BFO marked SET.  Here
sideband is "correctly" positioned in passband and BF0 is at the
suitable frequency. Move VXO pot. and that moves sideband across
passband. Now, just alter BF0 untill audio resolves.  Don't think
there is a need for automatic tracking of BFO when VXO is tuned.
Could arrange for that, you just derive BFO frequency from 2nd
mixer VXO.

Ought to study what Drake did.