Digital AM/FM Signal Generator

[peterw...@gmail.com]

I have come into a small, but totally unexpected windfall. As is the way of these things, it will be split into thirds, one to savings, one to my wife for spending on the house - as is her habit, and one to play with.

I am looking for a *SIMPLE* AM/FM/SW signal generator - need not be stereo, but I would like it to be digital. And, cheap.

I have two vintage devices that work just fine, but as I am gently approaching my "retirement" bench (still about 5 years out, but close enough) I am trying to reduce the clutter to fewer, but easier to use instruments.

I see devices in the many thousands of dollars - no need for that. Nor will I fuss about Chinese-Origin equipment in this case, as long as it is reliable and simple. The number of radios I actually align in a year, or FM tuners may be counted on both hands, or fewer. And of those, few of them are so far off as to require much effort - well, maybe that last Hallicrafters.

If this is too much to expect, what about a _REALLY_ good analog device, small, simple and solid-state? I am patient, and can wait as well. No rush whatsoever.

My budget is at/around US$320.

Along these lines - what about something like this?

https://www.ebay.com/itm/Leader-LSG-221B-Signal-Generator/232804889856?hash=item363440b500:g:qK4AAOSwVuNbGAML

Thanks in advance!


Peter Wieck
Melrose Park, PA

[Phil Allison]

pf...@aol.com wrote:

>
>
> My budget is at/around US$320.
>
> Along these lines - what about something like this?
>
> https://www.ebay.com/itm/Leader-LSG-221B-Signal-Generator/232804889856?hash=item363440b500:g:qK4AAOSwVuNbGAML

** That unit has a frequency range of 25MHz to 499MHz.

Specs are hard to find so I am not sure if it does wide FM ( +/- 75 kHz ).

Being "synthesised" makes it very stable, but so are all the broadcast stations.



..... Phil

[gghe...@gmail.com]

I like my Rigol DG1022. But only goes to 25 MHz.

https://www.rigolna.com/products/waveform-generators/dg1000/

100 MHz get's more expensive.

George H.

[peterw...@gmail.com]

What I am looking for is something that goes from Longwave through FM (US commercial band).

I would like it to be digital so that I could check it (once) with a proper counter, and the rely upon it until the next check. I am not calibrating NASA probes, but vintage AM/SW radios and (mostly) vintage FM tuners. I do have a FM-Stereo transmitter that will give me right/left only and Mono as well as fully frequency-agile, and vary from a few MW to about 50 MW, but if I have something that is really off, an independent SG would be helpful without requiring a lot of fuss.

Thanks again!

[Fox's Mercantile]

On 6/20/18 10:12 AM, pf...@aol.com wrote:
> What I am looking for is something that goes from Longwave
> through FM (US commercial band).

For 90% of what I do, I use an HP3325B
DC to 20 MHz (40Mhz on the rear panel)
Specifically because it's a direct entry via keyboard.
It does swept output as well.
Obviously, it will hand 10.7 MHz FM IFs.
You can modulate it with an external generator.



--
"I am a river to my people."
Jeff-1.0
WA6FWi
http:foxsmercantile.com

[Jeff Liebermann]

On Tue, 19 Jun 2018 11:00:26 -0700 (PDT), "pf...@aol.com"
<peterw...@gmail.com> wrote:


>My budget is at/around US$320.

Perhaps an HP 8640B generator. I have several and used to fix them.
Looks like the selling price is about $300 (including shipping) on
eBay:
<https://www.ebay.com/sch/i.html?_from=R40&_nkw=hp+8640b&_sacat=181992&Brand=HP&LH_Sold=1&LH_Complete=1&_sop=15>
Hmmm... Might go slightly over your limit, but methinks is with it.

Also, I've been able to find rather old service monitors, which I
consider more useful. Wavetek or SSI 3000A. I think I paid about
$300.

Gotta run...

--
Jeff Liebermann je...@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

[peterw...@gmail.com]

On Wednesday, June 20, 2018 at 12:14:41 PM UTC-4, Fox's Mercantile wrote:

Obviously, it will hand 10.7 MHz FM IFs.
You can modulate it with an external generator.

Um.... Um....

Meaning I will replace one (1) analog incumbent with two items. I am half-German, so I should find this appealing at some level, but it defeats the entire purpose of the exercise.

[peterw...@gmail.com]

On Wednesday, June 20, 2018 at 12:16:27 PM UTC-4, Jeff Liebermann wrote:

file:///C:/Users/pwieck/Downloads/8640B-Signal-Generator-Operating-and-Service-Manual-08640-90017.pdf Give me the manual-seems to do about everything I want.

There is one that would be $350 FOB 19027, and with a 30-day full guarantee. Looks very clean.

Thoughts?

[Fox's Mercantile]

On 6/20/18 11:52 AM, pf...@aol.com wrote:
> seems to do about everything I want.

About the HP8640, very nice piece, BUT the lower frequency
limit is 500 KHz.

[Jeff Liebermann]

On Wed, 20 Jun 2018 09:52:08 -0700 (PDT), "pf...@aol.com"
<peterw...@gmail.com> wrote:

>On Wednesday, June 20, 2018 at 12:16:27 PM UTC-4, Jeff Liebermann wrote:
>file:///C:/Users/pwieck/Downloads/8640B-Signal-Generator-Operating-and-Service-Manual-08640-90017.pdf

Ummm... that's the URL that points to your own c: drive.
There are downloadable HP8640 manuals all over the internet. However,
there are variations, packaging differences, options, and specialty
variations such as avionics or military versions. I think I have most
everything in PDF form and can snail mail a CD or arrange for you to
download them.

>Give me the manual-seems to do about everything I want.
>There is one that would be $350 FOB 19027, and with a 30-day full guarantee. Looks very clean.
>Thoughts?

My pitch line is that you can get some really nice used HP RF
generators for little money. The danger is that it might be too old
to get parts, or have some well hidden problem that you don't
immediately notice. A 30 day guarantee is fine, but you pay shipping
both ways. At 45.8 lbs (20.8 kg), that's not cheap.

There are a few potential mechanical problems. The big one is that
the bevel gears in the frequency range and modulation rotary switch
mess are cracked. Someone makes replacement gears using 3D printing,
but I haven't seen them. I would prefer brass. I'm tempted to make
some but would need to buy an indexer for the mill that I'm sharing.
Here's what it looks like:
<http://802.11junk.com/jeffl/pics/HP8640B/>
I can explain how I fixed the gears later. The good news is that once
the hole in the plastic gear is enlarged to fit the brass ring, and
glued, it stays fixed.

Here's a list of some other potential problems and fixes. Some of my
stuff is mixed in there, somewhere:
<http://www.ve7ca.net/TstH86.htm>
I have an HP8640b on the bench (or rather under the bench) that needs
a new RF output amplifier.

Check the options. Opt 1 and 2 involved the frequency ranges as
explained in:
<http://electronicsrevisited.com/html/hp_8640b.html>
Opt 4 is for avionics. You definitely want Opt 3, which is reverse
power protection. Without it, you run a good chance of accidentally
burning out the output attenuator. There's a 0.1A fuse in the N
connectors, but it's better to let Opt 3 protect the unit.

They don't make them like this any more:
<http://802.11junk.com/jeffl/pics/HP8640B/slides/hp8640b-fix.html>

However, I'm wondering if an HP8640B is what you really need. The
cavity oscillator is really low noise, which is what I need for SSB
noise and IMD measurements, but not much else. Perhaps a synthesized
generator might be better. I have an HP8656A generator:
<https://www.ebay.com/sch/i.html?_nkw=hp8656+-Philips>
Looks like the average selling price of $350 is over your limit:
<https://www.ebay.com/sch/i.html?_nkw=hp8656+-Philips&LH_Sold=1&LH_Complete=1>
Maybe haggling with the seller will drop the price into your range.

I have an HP8656A:
<http://802.11junk.com/jeffl/pics/home/slides/test-equip-mess.html>
Upper right, bottom of the pile on the shelf. Over the last 10(?)
years I had a few problem.
- The rear panel RF output connector is rather inconvenient.
- Calibrating the RF output, deviation, and AM modulation, were needed
after I screwed things up my not reading the instructions.
- The push buttons on the front panel tend to hit the front panel
bezel and stick. I'm sure I can fix it if I take it apart, but
haven't had the desire or time.

[Jeff Liebermann]

On Wed, 20 Jun 2018 16:42:08 -0500, Fox's Mercantile <jda...@att.net>
wrote:

>About the HP8640, very nice piece, BUT the lower frequency
>limit is 500 KHz.

The actual range is 450kHz to 550MHz.
It will go to 1024Mhz with the Opt 2 frequency doubler.

[Jeff Liebermann]

On Wed, 20 Jun 2018 09:52:08 -0700 (PDT), "pf...@aol.com"
<peterw...@gmail.com> wrote:

>Thoughts?

"Choosing an RF Signal Generator"
<http://www.ab4oj.com/test/sig_gen.html>

"HP-8640 Service: Hints & Kinks"
<http://www.ve7ca.net/TstH86.htm>

"The HP 8640B RF Signal Generator"
<https://www.ebay.co.uk/gds/The-HP-8640B-RF-Signal-Generator-/10000000010610658/g.html>

"Phase noise measurements microwave bricks compared to HP8640B,
HP8662A, and HP8656A"
<http://www.ke5fx.com/brick/brick.htm>

"HP 8640B Repair"
<http://jvgavila.com/hp8640b.htm>
<http://jvgavila.com/wb1.htm>

YouTube videos on HP8640B"
<https://www.youtube.com/results?search_query=hp8640B>

Gotta run (again)...

[Dave Platt]

>> seems to do about everything I want.
>
>About the HP8640, very nice piece, BUT the lower frequency
>limit is 500 KHz.

And, it's not really "digital" as the OP requested... it's an analog
cavity oscillator. The commercial versions (but not the military OPT
323 variant) do have a frequency-lock capability which will lock to
the crystal oscillator in the counter timebase, though... with this in
use its stability is quite good and I'm told the phase noise level is
excellent.

It's not quite within the OP's stated budget, but... I recently picked
up an 8648B for under $500, delivered. Very nice beast, covers 100
kHz up to 2 GHz, with AM and FM and phase modulation, and good enough
GPIB control to be useful. Even without the high-stability oscillator
option, it was within 1 PPM after a warmup, and it trimmed quite
nicely to within their recommendations (10 PPB) using my rubidium
standard as a reference.

[Dave Platt]

In article <k9olidteq0j79f2uo...@4ax.com>,

Jeff Liebermann <je...@cruzio.com> wrote:

>There are a few potential mechanical problems. The big one is that
>the bevel gears in the frequency range and modulation rotary switch
>mess are cracked. Someone makes replacement gears using 3D printing,
>but I haven't seen them. I would prefer brass.

There's a guy in India who has been making sets in brass... I'm in the
middle of installing a set now. He posts on the HP test-equipment
mailing list, and has some of the brass gears up on eBay.

The other issue I've run into, with that switch assembly, is that the
two white plastic parts (the shaft which holds the two small planetary
bevel gears, and the rectangular piece which clamps onto this shaft
and actuates the rear switch decks) are a bit dodgy. The tighten-it-
down clamp head on the rectangular piece was cracked, and even after
repair it isn't gripping the shaft well enough to prevent slipping.
And, the new brass gear which _should_ rotate freely on this shaft, is
rubbing and squeaking... either its hole is a bit under-sized, or the
plastic shaft isn't dimensionally stable after all of these years.
So, I have a bit of adjustment to do.

I'd love to replace both of these plastic bits with something more
stable (maybe a metal shaft with metal axles for the planetary bevel
gears, and a composite or Delrin assembly in back) and might try
fabricating something if I can't get the existing plastic parts back
into service.

[Jeff Liebermann]

On Wed, 20 Jun 2018 18:09:22 -0700, dpl...@coop.radagast.org (Dave
Platt) wrote:

>In article <k9olidteq0j79f2uo...@4ax.com>,
>Jeff Liebermann <je...@cruzio.com> wrote:
>
>>There are a few potential mechanical problems. The big one is that
>>the bevel gears in the frequency range and modulation rotary switch
>>mess are cracked. Someone makes replacement gears using 3D printing,
>>but I haven't seen them. I would prefer brass.

>There's a guy in India who has been making sets in brass... I'm in the
>middle of installing a set now. He posts on the HP test-equipment
>mailing list, and has some of the brass gears up on eBay.

<https://www.ebay.com/itm/Brass-Gears-for-the-HP-8640B-Signal-Generator/153024403695>

You should have taken my advice and glued the cracked gear back
together with epoxy. The trick is to temporarily put the gear back
together with a hose clamp or fixture. Then take a big rat tail file
and enlarge the hold in the plastic until the brass lock ring fits
easily in the hole. Or, you can use a proper reamer and a mill.
Remove the set screw and plug the hole with wax, or a longer screw
smeared with a little mold release (grease or Vaseline).

Leaving the gear halves clamped together, smear some 24 hr epoxy on
the knurled part of the brass lock ring, and let it dry overnight.
Remove the temporary clamp holding the gear together, remove the hole
plug, clean the threads, re-insert the set screw, and put it back into
the gear assembly.

I've done about 10 gears like this without experiencing any problems.
Well, not quite. The last one I did, I either got the mixture wrong
on the epoxy, or used some ancient tube of epoxy which refused to
harden. I had to scrape it off and glue it again, which worked.

Incidentally, while you have the gearbox apart, bend the other
non-broken gears a little to see if they are in any danger of breaking
in half. Most will break at the set screw with very little pressure.
Might as well have it break now, rather than while you're using the
generator.

If you still have the old gears, I'll be happy to glue them together
for you for the price of shipping and small donation to support my
decadent and lavish lifestyle.

>The other issue I've run into, with that switch assembly, is that the
>two white plastic parts (the shaft which holds the two small planetary
>bevel gears, and the rectangular piece which clamps onto this shaft
>and actuates the rear switch decks) are a bit dodgy. The tighten-it-
>down clamp head on the rectangular piece was cracked, and even after
>repair it isn't gripping the shaft well enough to prevent slipping.

I haven't seen that yet. I'll take a look at the two I have when the
bench is clear. That might explain some backlash in the shaft.

>And, the new brass gear which _should_ rotate freely on this shaft, is
>rubbing and squeaking... either its hole is a bit under-sized, or the
>plastic shaft isn't dimensionally stable after all of these years.
>So, I have a bit of adjustment to do.

I'm not thrilled with the machining quality shown in the photos. It's
adequate, but not great. If anything, the plastic shaft has SHRUNK
making a loose fit more likely than a tight fit. I can measure mine
if you need a comparison. I would bet the hole has burrs, shavings,
or is undersized.

>I'd love to replace both of these plastic bits with something more
>stable (maybe a metal shaft with metal axles for the planetary bevel
>gears, and a composite or Delrin assembly in back) and might try
>fabricating something if I can't get the existing plastic parts back
>into service.

Can you generate the G code needed to make the part? I'm helping
setup a CNC mill that can possibly make it.
<http://802.11junk.com/jeffl/pics/CNC-conversion/>
The catch is that we don't have a rotary indexer or gear cutter, so
all the teeth will need to be shaped with a small end mill. That's
slow and not fun. It would almost be easier to make a silicone rubber
mold and cast the gears from resin.

[Dave Platt]

In article <q36mid99k9epbqpk1...@4ax.com>,

Jeff Liebermann <je...@cruzio.com> wrote:

>You should have taken my advice and glued the cracked gear back
>together with epoxy.

I may well do that, to put my existing gears back in usable condition.

>I've done about 10 gears like this without experiencing any problems.
>Well, not quite. The last one I did, I either got the mixture wrong
>on the epoxy, or used some ancient tube of epoxy which refused to
>harden.

I've made a lot of use of G-Flex epoxy for jobs like this - it works
very well on most plastics, and also on metal.

>Incidentally, while you have the gearbox apart, bend the other
>non-broken gears a little to see if they are in any danger of breaking
>in half. Most will break at the set screw with very little pressure.
>Might as well have it break now, rather than while you're using the
>generator.

Both of the larger gears with set-screws are cracked, and will need
repair. The third large gear (the one which normally free-wheels on
the shaft) isn't cracked, and I don't think it has a set-screw hole,
but I suspect the nylon has probably shrunk and is putting the plastic
under tension/compression. Might be worth it to deliberately split it
between teeth, and give it the same file-and-epoxy treatment.

>>The other issue I've run into, with that switch assembly, is that the
>>two white plastic parts (the shaft which holds the two small planetary
>>bevel gears, and the rectangular piece which clamps onto this shaft
>>and actuates the rear switch decks) are a bit dodgy. The tighten-it-
>>down clamp head on the rectangular piece was cracked, and even after
>>repair it isn't gripping the shaft well enough to prevent slipping.
>
>I haven't seen that yet. I'll take a look at the two I have when the
>bench is clear. That might explain some backlash in the shaft.

Looking at pictures on the net, I suspect that at least part of this
problem exists only on the "old" switch-and-gear assemblies - the ones
made with real wafer switches. The "new" switch-and-gear assemblies
that use the PC-board-style switches with the plastic-disc-and-finger
contactors, may use a different type of shaft for the planetary gears;
the photos I've seen show the use of a metal tube with metal axles,
rather than plastic.

It's quite possible that the rear plastic part (the coupler) simply
doesn't exist at all in the "new" assembly, since the rear switch
section seems to be another of the PC-board type rather than a stack
of three wafers.

So, the whole "bad plastic tube and coupler" problem may not exist at
all on a lot of the 8640Bs.

>I'm not thrilled with the machining quality shown in the photos. It's
>adequate, but not great. If anything, the plastic shaft has SHRUNK
>making a loose fit more likely than a tight fit. I can measure mine
>if you need a comparison.

I'd appreciate that - thanks@

> I would bet the hole has burrs, shavings,
>or is undersized.

I think it may be both... that is, some burrs, possibly undersized
(although the other two did fit OK onto the metal shafts of the
switches), and maybe some changes in the plastic.

Here's what I got when I measured everything with a Miyutoyo digital
caliper:

Central rod .1120-.1125
Shaft ID .129
Shaft OD .2470 at the tail, as much as .2505 next to the axles.
Gear ID .2490 to .2500
Axles .124-.125
Planetary gear center .125

So, the shaft does seem to be a bit over-size where the gear is
supposed to rotate on it - possibly some damage there - I should be
able to smooth it down. I've already burnished the inside of the
gear to get rid of any burrs that might have been there.

I suspect that the two plastic planetary gears may have shrunk a bit
with age, just as the plastic parts of the three big gears did - I
need to open the holes up a bit to get them to rotate easily on the
axles.

>>I'd love to replace both of these plastic bits with something more
>>stable (maybe a metal shaft with metal axles for the planetary bevel
>>gears, and a composite or Delrin assembly in back) and might try
>>fabricating something if I can't get the existing plastic parts back
>>into service.

It looks as if I'd be able to use some fairly "stock" metal parts to
make the shaft... 1/4" OD fairly-thick-wall brass tube, and 1/8" brass
rod. The trickiest part would be mounting the 1/8" axle rods... I'd
probably have to drill slightly-undersized holes in the sides of the
tube, press-fit the axle rods into place, and then maybe silver-solder
them. Gotta talk to my machinist friend about this.

>Can you generate the G code needed to make the part?

Maybe after six months teaching myself how to use a CAD program :-)

> I'm helping
>setup a CNC mill that can possibly make it.
><http://802.11junk.com/jeffl/pics/CNC-conversion/>
>The catch is that we don't have a rotary indexer or gear cutter, so
>all the teeth will need to be shaped with a small end mill. That's
>slow and not fun. It would almost be easier to make a silicone rubber
>mold and cast the gears from resin.

I _think_ the gears I have will work out - either the plastic ones
(repaired) or the new brass ones I bought (maybe after some burnishing
or other tweaking). It's the center tube and planetary-gear drive
that I'd need, and I'm not sure CNC milling would be the way to go.

I did find a supplier who sells Delrin rod with a slightly-oversized
1/4" diameter and a slightly-undersized 1/8" central opening. This
might work, if I drilled out the inside to a bit more than 1/8", and
then machined down the OD to .125". Unfortunately I haven't found
anybody who sells this particular variant in small quantities.

[M Philbrook]

In article <39vkid50vpmlm75ue...@4ax.com>,
je...@cruzio.com says...

>
> On Tue, 19 Jun 2018 11:00:26 -0700 (PDT), "pf...@aol.com"
> <peterw...@gmail.com> wrote:
>
>
> >My budget is at/around US$320.
>
> Perhaps an HP 8640B generator. I have several and used to fix them.
> Looks like the selling price is about $300 (including shipping) on
> eBay:
> <https://www.ebay.com/sch/i.html?_from=R40&_nkw=hp+8640b&_sacat=181992&Brand=HP&LH_Sold=1&LH_Complete=1&_sop=15>
> Hmmm... Might go slightly over your limit, but methinks is with it.
>
> Also, I've been able to find rather old service monitors, which I
> consider more useful. Wavetek or SSI 3000A. I think I paid about
> $300.
>
> Gotta run...

I have two them with the 1G option, love'm