-
#25 Reply
Posted by
Jidis
on 21 Aug, 2017 22:01
-
Thanks again bd139! Got one on the way from them to see what it does, along with a handful of other junk. Wasn't aware of that place. Looks like a good assortment of stuff they have, and I hate having to pay UPS shipping rates for two bucks worth of junk from Mouser.
Looking at one of the chips here, mine has F0918 printed on it, so I guess that's a 2009, like one of the ones in the video Mick linked to. I didn't realize people kept making them that recently. For some reason I was thinking it was something that died way back before 2000.
Take Care
PS- I kick myself every time I see one of those 1054 scopes from that video. Missed that bus by about six months (1052E here).
-
#26 Reply
Posted by
bd139
on 21 Aug, 2017 23:59
-
Yeah they're really good. Also they mark it up as $10 commercial samples always so you never get screwed for import
I think Exar stopped selling them in about 2001.
1054Z here as well bought about three weeks ago. Nice bit of kit. I'm sure something better will come out in a month and make me feel the same
-
#27 Reply
Posted by
rdl
on 23 Aug, 2017 06:13
-
I'm pretty sure the XR2206 was discontinued in 2011
-
#28 Reply
Posted by
bd139
on 23 Aug, 2017 10:42
-
I think that's when the last fab run stock ran out and they decided there wasn't enough cash to do another one so pulled it off the market. The genuine ones I've had were all latest 01 date code going as far back as 2009. It hasn't exactly had a good time with DDS ICs and microcontrollers with DSPs in them for less cost on the scene.
-
-
The original Exar is a good part, but long gone. A few years ago I got some from Jameco that obviously weren't good copies of the original. They generally work OK, but won't meet the sweep specs for a real Exar part, and only come close for one specific supply voltage. I'd be wary of using them in any circuit where a wide range DC signal is used for frequency control.
-
-
I was sure I had a NOS part in my bin to give away, turns out it's a 556! Who knew Exar made 556s?
-
#31 Reply
Posted by
rdl
on 23 Aug, 2017 17:54
-
I remembered I bought two from Mouser around 2004 or 2005.
I can only find one.
Now I will probably waste the next hour or so looking for the other.
Sigh.
-
#32 Reply
Posted by
Jidis
on 23 Aug, 2017 18:54
-
I wonder why mine have an "F" in front of the date code?
Fake?
-
#33 Reply
Posted by
blackdog
on 23 Aug, 2017 19:17
-
Hi Group,
About four years ago i designed a XR2206 generator for a peace of measuring test set, bus its also usable stand alone.
But first a picture of the schematic.
It uses 3 frequency ranges, so designed that you can quickly sweep between 10HZ and 50KHz.
The output impedance is a nice 50 Ohm (Do NOT load it with 50 OHM) it is a one site terminated setup.
It hase a LOG range amplitude control potmeter and one switch so you wil have a 0.3mV to 3.4V RMS range.
This design has olmost no overshoot on the top of the Sinus wave!
The squarewave is nice without abberations and nog load dependend
Do NOT change the way i designed the frequency control, you wil be sad afterwards
There was a lot of design time in this relative "simpel" generator, but it gets everything out of de XR2206 and olso a stable output output impedance for Sinus and Square wave.
On a Dutch forum i have a article about generator, have a look, also this desing is mentioned, you can always use Google translate...
https://www.circuitsonline.net/forum/view/112672/1/0Kind regards,
Blackdog
-
#34 Reply
Posted by
gtrc
on 23 Aug, 2017 21:39
-
I found this old topic about the XR2206 function generator and wanted to share my experience on a project. I made a device that continuously recorded 'environmental data.' The XR2206 was the audio interface to a PC sound card demodulator. The system plotted a continuous time vs signal level graph. The computer produced and saved data for 30 minute recordings of information. Really cool system. Put a few hundred hours into the project, including the programming for the computer and the inst amplifier and conditioning circuit for the system. Long and the short of it was that after 6 weeks of continuous recording, in the middle of one 30 minute plot, the frequency produced by the XR2206 started increacing by itself. In increased about 20 hz in each half hour, or possibly more. This continued until the program could no longer plot any trace on the screen. I changed to another XR2206 and 6 weeks later the same failure occured. Bought more from a different supplier trying to get a different lot, and same thing happened again. Very disheartening to say the least. Called XR factory and they said they would look into it. Never heard anything back. My circuit was an exact duplicate of the datasheet. I went from being thrilled at having created a living thing, to complete desolation at having wasted so much time. And the chip was tunable to get very low harmonic distortion of the sine wave. I'm sure it was just a weak trace inside or some little part that slowly eroded away. Anyway, I would not recomend the chip for continuous duty. 6 weeks gives you a lot of hours of operating time, and maybe intermittent use preserves the chip. I think the chip should be redesigned to fix that problem. Then it would be a nice device for experimenters.
-
#35 Reply
Posted by
blackdog
on 24 Aug, 2017 07:57
-
Hi gtrc,
On witch frequency was the XR2206 used?
If it was 100Hz then 20Hz is a lot, if it was 20 Hz om 15KHz it is not zo much...
Some tips
Use stable components!
Do not load the XR2206
Use a stable power supply.
Short wiring, groundplane on the circuit boards
Kind regards
Blackdog
-
#36 Reply
Posted by
Jidis
on 24 Aug, 2017 18:59
-
Blackdog,
Wish I had looked into yours when I did the one here with the floating ground issue. It's already got an enclosure and panels, but I may still think about replacing it if it doesn't perform much better when my "hopefully authentic" chip shows up.
If mine could do anything remotely close to that distortion spec you got, that would be well worth it. Out of curiosity, does it get much noisier down at 1kHz?
Thanks!
-
#37 Reply
Posted by
blackdog
on 24 Aug, 2017 20:00
-
Hi Jidis,
The distortion spec is on most of the frequency range below 0.8% and you can tune the distortion in de audio range for a big part to about 0.5 a point 0.6%.
Depending on de XR2206 you have at the moment.
I cant remember reail noise issues and just measured one of the generators.
It is not difficult to get the distortion below 0.6%, but if you make the sweep range to big say from 500Hz to 35Khz, you never get this low distortion figure over the whole range.
Look @ the pictures below, the distortion was of no big inportance for the use of this generator (Build in a old Cico Pix firewall housing)
How one of the generators is build.
1KHZ distortion, its a little high but thats because i used a big range for this generator.
Analyzing the 1KHZ distortion.
10Khz, and now the distortion is better because we are in the middle of the frequency range.
Analyzing the 10KHZ distortion, now we have a neat distortion figure for a XR2206
Kind regards,
Blackdog
-
#38 Reply
Posted by
Jidis
on 24 Aug, 2017 20:50
-
Thanks Blackdog!
I think one of the replies I skimmed over at that circuitsonline thread gave me the impression that you were actually getting those <0.0x% figures from a 2206 circuit, though <1% would still be a heck of an improvement over what I was able to get on mine. I've still got those Twin-T things if the goal is to try to check something's distortion at 1k anyway.
Take Care
-
#39 Reply
Posted by
Jidis
on 27 Aug, 2017 03:07
-
bd139-
Unfortunately, it's a no-go on that Exar chip from Tayda (got here today). I was sort of expecting it when I saw the same "F0918" code on theirs. It's weird that they're the exact same number, being as mine were bought on eBay over two years ago. If I had any sense I would have specified when I ordered that I was looking for something other than that (ideally older). They had a box on the order page for additional instructions.
Either way, thanks again for the tip on them. Only four days to get it here in VA. It says it came from Colorado or something. Packaging is nice too with nice bold printed labels on all the bags. All my Mouser stuff used some sort of gag gift disappearing ink, so I've got drawers full of parts now with no labels.
If you or anyone here ever gets a chance to check one of the older ones with that same opamp floating ground I used, I'd be very interested to know if there's something that works with it. The site that was on (Nuxie?) had premade boards and kits, which I doubt they'd do if nobody ever checked it. The floating ground part is pretty simple (opamp,3 resistors, and a cap or something). I guess it would do the same no matter what the rest of your XR2206 circuit looked like, but note they do have the amplitude pot using the opamp ground too.
Take Care
-
-
Bring back memories, i had an xr2206 kit, very well made, i'm trying to find the schematics ... had symetrical supply with push pull transistors for outputs, could go at 10v pk pk and drive 50 ohms loads, 1 v pk to pk at 1 mhz ...
Changed to a max038 design, it was fun at the time ordering free samples from Maxim lolll
-
#41 Reply
Posted by
bd139
on 27 Aug, 2017 09:13
-
bd139-
Unfortunately, it's a no-go on that Exar chip from Tayda (got here today). I was sort of expecting it when I saw the same "F0918" code on theirs. It's weird that they're the exact same number, being as mine were bought on eBay over two years ago. If I had any sense I would have specified when I ordered that I was looking for something other than that (ideally older). They had a box on the order page for additional instructions.
Either way, thanks again for the tip on them. Only four days to get it here in VA. It says it came from Colorado or something. Packaging is nice too with nice bold printed labels on all the bags. All my Mouser stuff used some sort of gag gift disappearing ink, so I've got drawers full of parts now with no labels.
If you or anyone here ever gets a chance to check one of the older ones with that same opamp floating ground I used, I'd be very interested to know if there's something that works with it. The site that was on (Nuxie?) had premade boards and kits, which I doubt they'd do if nobody ever checked it. The floating ground part is pretty simple (opamp,3 resistors, and a cap or something). I guess it would do the same no matter what the rest of your XR2206 circuit looked like, but note they do have the amplitude pot using the opamp ground too.
Take Care
That's a bit annoying. I will order one as well and rig up something and see where I get. Will post back in a few days.
-
#42 Reply
Posted by
Jidis
on 27 Aug, 2017 21:00
-
I will order one as well and rig up something and see where I get. Will post back in a few days.
Don't go to any trouble on it. I was under the impression that you still had the old ones you got from them laying around somewhere. I'd imagine you'd get the same thing I got if you ordered one now.
Thanks!
-
#43 Reply
Posted by
Jidis
on 31 Aug, 2017 01:03
-
This is nerve racking. I got a reply from Tayda that they only import the 0918 ones because several times 0919's were rejected after failing tests there. Both of those seem to be all over eBay with 0918's probably being more common.
What exactly is that number anyhow? I'm figuring the "09" would be year, but "18" obviously isn't month. I also saw ones on eBay with numbers (1435) which made no sense if it was discontinued in 2011. Speaking of which, optimistically assuming a bunch of them are legit, why is there this whole slew of them with these same two numbers? Shouldn't the factory have been pumping them out all along until they stopped in '11, with numbers from just about every year prior, or are all the ones we get now just high functioning fakes where the counterfeiters just didn't bother to print different numbers on them?
Thanks!
-
#44 Reply
Posted by
JoeO
on 31 Aug, 2017 02:57
-
This is nerve racking. I got a reply from Tayda that they only import the 0918 ones because several times 0919's were rejected after failing tests there. Both of those seem to be all over eBay with 0918's probably being more common.
What exactly is that number anyhow? I'm figuring the "09" would be year, but "18" obviously isn't month. I also saw ones on eBay with numbers (1435) which made no sense if it was discontinued in 2011. Speaking of which, optimistically assuming a bunch of them are legit, why is there this whole slew of them with these same two numbers? Shouldn't the factory have been pumping them out all along until they stopped in '11, with numbers from just about every year prior, or are all the ones we get now just high functioning fakes where the counterfeiters just didn't bother to print different numbers on them?
Thanks!
The second number is the week. So 1435 is the 35th week of 2014.
-
#45 Reply
Posted by
Jidis
on 31 Aug, 2017 03:34
-
The second number is the week. So 1435 is the 35th week of 2014.
Thanks Joe! That makes sense, though it still doesn't explain how they were making them three years after Exar stopped on that particular one though.
-
#46 Reply
Posted by
bd139
on 31 Aug, 2017 14:03
-
I will order one as well and rig up something and see where I get. Will post back in a few days.
Don't go to any trouble on it. I was under the impression that you still had the old ones you got from them laying around somewhere. I'd imagine you'd get the same thing I got if you ordered one now.
Thanks!
I have a use case for one (FSK modulator) and a whole pile of stuff to order ($30 worth which is a lot of stuff for Tayda!) so no problems running a couple of tests first. It left Thailand on Tuesday so will see what happens when it arrives.
-
#47 Reply
Posted by
LukeW
on 01 Sep, 2017 11:55
-
They're fun, interesting chips from before the "do everything with microcontrollers" generation.
It's a shame there's no reliable way to get them any more.
-
#48 Reply
Posted by
bd139
on 01 Sep, 2017 12:41
-
You can make one pretty easily with discrete components to be honest. All you need is a current source, sink, high Z follower, window comparator and diode switch with a diode based shaper. I'll knock one up at some point. That's how all the commercial (Tektronix / HP etc) generators worked.