0051 - (Fail?) Improving the JDS6600 Part 2: The Time Base
Вставка
- Опубліковано 25 сер 2024
- In this 2 part series we are going to try and improve this JDS6600-60 by eliminating some of the problems associated with the power supply and the time base. In Part 2 we replace the stock oscillator with a TCXO. However we find out there are other issues...
Help support the channel directly: paypal.me/UniB...
Buying from AliExpress? Use this link to help the channel and it won't cost you a penny more!
AliExpress Affiliate Link: s.click.aliexp...
JDS6600 AWG: s.click.aliexp...
![Every Maker should have...[Pt.37]...a Rubidium Frequency Standard](http://i.ytimg.com/vi/zW5ffFuEQsw/mqdefault.jpg)
Solid improvement; kudos! I liked your custom board and socket. I no longer buy any four-sided sockets. I just buy socket strips and make my own sockets. I do not have to pay for, or store, a selection of sockets and I always have a socket with the correct pin count on hand (grin.) I like the SamTec brand (Mouser has them.) I use model# SSA-132-W-T 99% of the time. The big, square, long legs are very adaptable, for creative applications, and they give a super-strong mechanical connection that holds up well to multiple chip exchanges. I use model# SSA-132-S-T, when I am too lazy to trim the pins. I also like their high-temperature header pins; model# HTSW-150-29-T-S. No worries with the extra dwell time needed to seat a header. No, I don't have any relationship with SamTec; I just think they make a good product.
Great tip, thanks for mentioning it. I plan to do the same, once I use up my 100's of sockets ..
Those square pin ones make good contact with IC pins? I was thinking the round machined ones would be good.
@@uni-byte Yes, very good contact. I very much like the square ones. So much that I have never tried the round ones, even though I have some. I am so satisfied that I do not plan to try the round ones. Just no need, IMHO.
@@t1d100 Cool, thanks for the update. I'll have to give them a try.
Nice job. I'm looking to put a better ref in my apparently similar cheap function gen. I'll have to find out what its clock ref is. 10 MHz would be ideal since I bought some of those $3 10 MHz Xtal ovens. But if I have to I'll add a phased lock loop circuit to send the unit whatever freq it needs. I like accuracy, but I can't always have everything tied to a GPS locked ref on my work-bench. Sometimes I grab a box of small equipment and take it to the local maker-space, so accurate portable gear is nice.
Sounds like a worthwhile project. Let us know how it turns out.
Have you looked under that little foil seal on the crystal to see if it has a trimpot?
Yeah. As I mentioned I suspected in the video it was only adjustable by about 10Hz, but the wrong way. So I just put it back to the 1.5Hz over the 24MHz. These TCXOs tend to be pretty accurate so it's not likely I would ever find one that is off by 148Hz, which is what I'd need to correct for the ~62Hz error @ 10MHz. But maybe a regular oscillator might be. Have to see if I can pick up a collection of them for cheap and try them out. Of course then I loose the stability of the TCXO.
@@uni-byte Yeah I think I'd just live with your current solution then. Bit weird the generation is so off. If you try some direct divisions of 24Mhz does it get any better? Pure sinewave at 1Mhz for example.
You could even think of seeing if there's an email contact for the manufacturer (Juntek)? I've seen there's quite a few trimpots on the main board of the unit, it would be a miracle if one of them was to fine-tune frequency...
Great upgrade to an already impressive device for the price.
I wonder if the frequency error after replacing the crystal is the result of however they did the frequency synthesis - maybe the multiplier/divider logic is a bit limited to keep internal frequencies and costs low. Wonder if other output frequency settings are more accurate?
Question - the "big" issue I find with JDS6600-60 is that it is so light it doesn't stay in place well on the workbench. I'm guessing the added weight of the power supply helps, but the thicker power cord probably makes the issue worse. Based on your experience with the power supply addition, was that enough additional weight to improve the mechanical handling of the device? Or would adding something like a steel plate inside for additional weight be of use?
Thanks for the great video!
I'm pretty sure you are right, the frequency error is in the synthesis. The power supply does add a little weight, but not much. I put little silicone feet on it to stop the tail (power cord) from wagging it. I also use a very supple power cord that I found in bulk at a nearby surplus shop. I picked up about 10 of those. You could also install some additional weight in it too. That would help.
@@uni-byte Thanks for the reply! This winter I might do the modifications and add some weight as well. Good thing its almost empty inside.
@@ElectromagneticVideos 👍 Let us know how it goes.
@@uni-byte Definitely!
Great job! I would really like to see, if posible, how you adjust the oscillator frequency. Thanks in advance!
There is a little hole in the can (covered by a little sticker). In there is a small variable capacitor. I have one not in a device in use that I can dig up and do a short on it. It's not accurate to begin with and the amount of adjustment is tiny. In the 10s of Hz.
Have a look at this video: ua-cam.com/video/kmqh8Szhx24/v-deo.html
I explain how it's done.
I guess the manufacturer knows that the cheap crystals are a bit off, and has added code to compensate for a 64Hz "average offset".
Could be, but they'd have to select for crystals that were on the higher side.
@@uni-byte Or, their batch of crystals were known to be a bit fast.
@@Batmule Except mine... It was slow by about 115Hz.
Yes, exactly. You need to input not 24.0 MHz but with a little offset in the opposite direction.
6:04 HAHAHAHA You can hear the disappointment in your voice....😂😂😂 Totally NOT what you expected....
No, I had hoped for a bit more ..
First time I've seen someone do this mod. Nice try. Is there an updated firmware for it?
Thanks.
I went over to the manufacturers site but there is no updated firmware for it. Maybe something out in the user community. I'll be taking a look.
@@uni-byte Let us know. It might come in handy some day.
@@user-lv7xv2sk7j Roger, WILCO.
Where did you get the daughter board?
Thanks for the video! I wonder if they make programmable TCXO's? Probably REALLY expensive.
I had it made up at PCBWay. Just real simple quick design. I guess you could make a really good programmable frequency reference with once of those $3 OCXOs and a PLL. Might not be too expensive but it would be pretty big and consume a lot of power. Or a TCXO and PLL for less power, but there does not seem to be a good supply of cheap surplus TCXOs.
@@uni-byte Thanks! Good ideas for when I do mine.
I'm thinking that I may add a buffer and a switch so that I can use an external source too (for my GPDSO).
@@button-puncher MY GPSDO is strictly 10MHz. I'd still have to conjure up a PLL.to get the 24MHz
@@uni-byte If you do, post up a how-to. I'd love to see it. That way I can just leave my GPSDO set to 10MHz for my frequency counter.
BTW, I've got the Leo Bodnar mini GPSDO. It's great! I highly recommend. I've got the one that only goes down to 400Hz. He has a new one that goes down to 1Hz. The receiver is sensitive enough to work near a basement window. I was really surprised by that.
@@uni-byte I just learned about the QRP Labs ProgRock2. You've GOT to check it out. It has a 0.25PM TCXO, accepts a 1PPS input, GPS option, and can be set to any needed frequency. So you can get your generator's frequency spot on by adjusting it slightly off 24MHz. That's the route that I'm going to take with my JDS6600.
I found your videos, trying to figure out the reason why mine has a hard time powering up. I push the button hear a beep and blank screen, if I keep depress and do a quick push again it comes on. Feels like hardware but not sure anyway great videos, might do the upgrade, the oscillator scares me a bit.
Thanks. So if you just push the button and let go it will beep but won't start up? The oscillator upgrade is not really worth it as most the error was in the firmware. The power supply is a really good upgrade though.
@@uni-byte Ok I had to do it several times to get the working protocol. If I just depress it I get the beep and blank. If I depress it and quickly press it again I get the main screen but it wont finish goes to blank. The third and final way, I depress it, quickly press again as the main screen is loading and the press again instantly then release loads up and then it beeps. I hope thats clear, I left it on for 8 hrs and it was working when I got back, All the functions work, the menu as well,
Forgive me for not understanding, but how can you replace the 10 MHz crystal with a 24 MHz crystal and the unit still works as advertised? (I re-watched the video a few times but I'm obviously missing something :/ )
The original oscillator was also 24MHz. If you look at the video at about 2:00 minutes in, I put the frequency counter on the little internal oscillator and it measures at 23,999,864 Hz.
Ah thank you... clearly age is not treating me well :D. I do like watching videos of people improving their gear though. I shall be more attentive in future.