Products > Test Equipment

Conversion of 500MHz TDS744A to 1GHz TDS784A

<< < (8/24) > >>

timb:
So, last night I took the plunge and did the conversion on my newly acquired TDS 754C.

Initially, all resistors *except* R1064 were populated:



To convert it to a TDS 784C, I removed R1061-63 and populated R1064:



Then I carefully removed the four capacitors, cleaned the area with foam swabs and isopropyl alcohol and reassembled the scope. She booted right up as a TDS 784C!

I hooked my TG 501 Time Mark Generator up, set it to 1ns markers and turned the scope down to the 200ps setting:



So far so good! Let's see how the signal looks in normal mode:



Now, some of the weirdness may be on the TG 501's end, I'm just not certain. (It came to me in pieces and I was never able to calibrate the 1ns output, because until now I didn't have a scope capable of it).

Anyway, I'm pretty happy with the conversion! I haven't calibrated it yet (because I don't have a signal generator that will go that high), but for now it's alright. At least I can get a good idea of higher frequency stuff now, plus there seems to be no change on the 2GS/s or lower ranges, so I can still rely on them and know they're accurate. (I assume this is because the scope already had the cal data up to 500MHz.)

Now to find a proper signal generator to do the calibration! Maybe I'll get lucky and find a SG 504 on the cheap; people want ridiculous amounts for them on eBay, even without the leveling head (which the unit is useless without).

rx8pilot:
Way to go Tim - I was always too scared to try the conversion since I had no real backup option. Now your 750Mhz active probes are not enough and you will be wanting faster ones. The cycle never ends.  :-DD

timb:

--- Quote from: rx8pilot on August 11, 2016, 12:56:58 am ---Way to go Tim - I was always too scared to try the conversion since I had no real backup option. Now your 750Mhz active probes are not enough and you will be wanting faster ones. The cycle never ends.  :-DD

--- End quote ---

Thanks! I actually got use my hot tweezers to remove the resistors. I bought them like two years ago and they've just sat in my soldering box... Worked perfectly though, no pulled traces or marks on the board!

I hadn't even thought of my probes... Better update my eBay searches!

Speaking of which, I just scored an SG 504 for $100 (the guy had it listed for $800)! No leveling head, but I did find schematics and a PCB layout for a DIY replacement. (There's even a guy on eBay who builds and calibrates them to 0.1%, which is ten times better than the original Tek head! He wants $120 though, so I'll have a crack at building one myself first.)

rx8pilot:
I have been hoping to get my hands on one of these fast pulse generators. Designed by esteemed forum member based on the Jim Williams design but with some clever updates.

I would love to get a kit, or gerbers, or a built unit, whatever. Obviously you need a proper signal generator first to calibrate, but it could be used to see how the scope responds to a fast edge. Since it only samples at 2Gs/s - you are right at the Nyquist at 1Ghz where the front end starts to roll off.

Jay_Diddy_B:
Hi,

Part of the calibration process is to adjust the timing between the samplers. The samplers are interleaved at the high sampling rates. When you do this the waveforms become a lot smoother. This a picture of mine after the calibration:



The scope use Equivalent Time sampling, indicated by the ET on the screen. Different parts of the waveform are acquired on different acquisitions. This fine for repetitive waveforms, but no use for single shot events.

There are a couple of Avalanche pulse generators on this forum. There is one designed by Free Electron.
This uses the 2N2369 transistor.

That thread can be found here:

https://www.eevblog.com/forum/blog/eevblog-306-jim-williams-pulse-generator/


I designed one that uses a BFR505 transistor which is significantly faster.

The thread starts here: https://www.eevblog.com/forum/projects/transmission-line-avalanche-pulse-generator/



I got to this level of performance:



This is measured with a 20 GHz scope.


This is what the risetime of my 744A looks like after conversion and calibration:






You need a waveform with a flat top to measure the risetime. If you don't you can get very optimistic results.

Good luck !!

Jay_Diddy_B


Navigation

[0] Message Index

[#] Next page

[*] Previous page

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod