changing the rigol DS1052E to DS1102E using USB , the dummy guide

[quadphile]

Hi, I send my thanks to all those involved so far!

In response to this bandwidth discussion I was wondering why a sinewave was not pursued first instead of a pulse generator.  The thought being to run the scope in the undersampling mode to get the "25Gs/s Equivalent Sample Rate", pushing the digital limitations far enough out to be no longer of concern (I'd hope!).  And if jitter is an issue in this mode then I guess you could do the same test but with normal sampling.

Then, sweep it with an RF source and spectrum analyzer for verification of input level consistency.  The sweep provides the -3dB point(s), and hence the bandwidth.  An underlying thought I had was that maybe these 50MHz scopes had poor flatness (level accuracy) from 50MHz to 100Mhz+.

I'm told I only know enough to be dangerous,    Thanks again, and hopefully I'm being helpful!

[clonecrp]

Ok Then  !  On to fixing the power supply etc (nice screen on this thing  ) ... =  Hantek - Tekway - DSO hack - get 200MHz bw for free 

Thanks !

[Bored@Work]

In response to this bandwidth discussion I was wondering why a sinewave was not pursued first instead of a pulse generator.

The pulse generator with 350 ps rise time (that is 350 * 10^-12 s) can be build by amateurs on a weekend for very little money (maybe 10 USD). It works, and is fun to do.

It works for measuring the bandwidth, because there is a rule of thumb formula, linking displayed risetime with oscilloscope bandwidth, as long as the generator's risetime is much better than the oscilloscope's one. With one simple mesurement one can see the oscilloscope's bandwidth.

Building or buying a 100 MHz or 200 MHz sine wave generator is a completely different game.

Then, ...  with an ... spectrum analyzer

More money down the drain for what an amateur can do with 10 USD and one look at the oscilloscope's display.

[tazman171]

Hi there.  New to the forums here but have been reading.  I just received my RIGOL 1052E and it is now a 1102E!  I bought it from Saelig for $399 including shipping and tax plus it cam with a carrying case.

Awesome info in here for doing this.

I did the upgrade via the USB method using shafri's rigupgr.exe program.  Took 15 minutes tops, thank you for all of the info here!    Couple of pics showing the software and hardware info before and after:

Before:



After:



Once again, great info and guides, thank you.

[clonecrp]

BoredAtWork

I looked @ the pulse generator circuit in the application note quite some tiime ago ... just never got around to it  (printed it out - thick note ! ) .... but I think you've changed my mind  - Looks like next weekend  .

A new Rigol convert ! Slick USB upgrade huh tazman171 ?

Doug

[torch]

Sorry, I said App Notes, but it was actually the datasheet (http://cds.linear.com/docs/Datasheet/1073fa.pdf) that helped me out. The IceNINE circuit is a copy of the App Notes circuit. The datasheet circuit is functionally similar, but with different values for the voltage supply caps. The first specify 0.47uF caps, the datasheet actually specifies 0.1uF caps. It made a real difference changing the one that is connected to the inductor to the lower value cap (the other two didn't seem to make as much difference). With the wrong caps, the voltage into the transistor was far too low -- something like 20v, IIRC. With the correct value caps, the voltage rises nicely above 90v, which is critical to get the sought-after pulse. At least, this is what happened in my case. YMMV.

One thing the app notes pointed out was the importance of keeping the leads as short as possible. I had much less ringing when I clipped the leads off so they were just long enough to reach the holes in the breadboard. Now that I have the circuit working, I'll try soldering everything direct to the BNC connector, as the app notes suggest. 

If I understand the principle correctly and assuming I am actually generating a pulse with a 350ps rise time, then the actual bandwidth of my particular scope with the supplied probes should be (0.34/2.12)*1,000 = ~160MHz. Not bad for a 50MHz price!

[tekfan]

Just for the info

The pulse generator actually produces pulses considerably less than 350ps. I've measured it with a 1GHz scope and I get around 310ps. Really quite amazing.

This has to be one of the simplest and highest performance builds in a long time.

[clonecrp]

torch & tekfan

Differences understood .. Thanks guys for the tip(s)  ... now it looks like I need to find some samples  ... Chip, Diode & Transistor ...

On a misson now .... & will let you know of success/failure ) ... no 1GHz scope here 
Thanks!
Doug

[clonecrp]

Back Again ! ... What chip version yeilds best success  ?

ie.: 

LT1073CN8
LT1073CN8-5
LT1073CN8-12 
LT1073CS8
LT1073CN8-5
LT1073CN8-12  etc...

[torch]

Don't use the -12 or -5 versions! They are pre-configured with internal resistors that set the output voltage to 12v and 5v, respectively, so you cannot get the required 90v (Don't Ask Me How I Know This.tm) (BTW: anyone have a project that requires 12v from a 1.5v battery? I have an LT1073CN8-12 I can let go real cheap...  )

The N8 package is an 8-pin DIP. The S8 package is an 8-lead narrow surface mount. I prefer the DIP.

[clonecrp]

Are you sure on part selection? Looks like the 90Volts is derrived by stacking off pin SW1 =>  http://www.i9t.net/fast-pulse/fast-pulse.html

So you are saying the winner should be => LT1073CN8    Kerrect?

[jahonen]

Pulse is too short for narrow bandwidth scope. Suitable piece of coax can be used to lengthen the output pulse without affecting the pulse amplitude or rise time, see my measurements here.

Regards,
Janne

[Bored@Work]

Back Again ! ... What chip version yeilds best success  ?

ie.: 

LT1073CN8
LT1073CN8-5
LT1073CN8-12 
LT1073CS8
LT1073CN8-5
LT1073CN8-12  etc...

As I have confessed previously, I didn't even use an LT part in my version, because I didn't have one at hand. Many boost converter ICs and schematics should work, as long as you can get the final output to 90V, without killing the IC. And as long as you can get the switcher working with the low load (see the 1 M? resistor).

Linear's original version uses 45V with a voltage doubler, because the LT1073 can't directly produce 90V. You could, for example, use a 30V IC with a voltage tripler (one diode stage more). LT also published other versions where they used an LT IC which could directly generate 90 V (some converter for CFL). There is nothing that really requires the LT1073. It is of course nice it can work from a single 1.5V battery, and you might need more batteries when using other ICs.

Just make sure you don't kill yourself. 90 V is already officially in the "is dangerous" zone. If you mess up your SMPS and accidentally create an even higher output voltage ...

(Dave: Can we get a "Zapped", "Grilled", "Fried" or "Blown up" emoticon?)

[tekfan]

Believe me the single AA battery operation is totally worth it. It uses so little current that you won't have to replace the battery in a long time.

The capacitors store a very small amount of charge so there's nothing to worry if you accidentaly touch the caps. You might feel a small spike but nothing serious. I've touched it several times and I just say oops every time.

I used 0.47uF caps rated at 63 volts. I wonder how long they will last. Well atleast they're good quality caps.

[torch]

So you are saying the winner should be => LT1073CN8    Kerrect?

That is what I used, yes.

I realize that the 90v is not directly output from the chip itself, but the circuit in total as provided cannot achieve the 90v with the -12 version and I don't know enough (yet?) to redesign the circuit.

[alm]

In response to this bandwidth discussion I was wondering why a sinewave was not pursued first instead of a pulse generator.

The pulse generator with 350 ps rise time (that is 350 * 10^-12 s) can be build by amateurs on a weekend for very little money (maybe 10 USD). It works, and is fun to do.

It works for measuring the bandwidth, because there is a rule of thumb formula, linking displayed risetime with oscilloscope bandwidth, as long as the generator's risetime is much better than the oscilloscope's one. With one simple mesurement one can see the oscilloscope's bandwidth.

Yep, for every complex problem, there is a solution that is simple, neat and wrong. But I'm sure it's fun to see those high bandwidth numbers after spending a few hours (or less) building the AN-47 pulser. A correct solution (which involves a charge line, see Janne's post for example) isn't that much more complex, though.

[torch]

Please expand on what is "wrong" and how to correct it.

[flyingfrancisco]

Hello everyone
I just statred +to nod my DS1052eE from version 02.05. SP2 and I have hit a wall. I have installed the 02.02. SP2 but can't connect to the scope with UltraScope. When I try to connect it asks me to choose from  ASRL1-INSTR or ASRL10-INSTR but either one doesn't work. I get the error "Can't Read from Device" and when I click ok I get "Device not found" I have uninstalled ULtrascope that came with cd and uninstalled "Visa462runtime.exe" reinstalled them and get the same errors. I did not get a request to install drivers. What am I doing wrong.
Thx in advance to everyone

flyingfrancisco

[alm]

Wrong: if the pulse duration is in the same order of magnitude (or faster) as the scope's rise time, the falling edge appears at the input when the scope's output signal is still rising. This interferes with the rising edge, and distorts the pulse. This will result in a shortened pulse, if the rise time is calculated from 10% to 90% of the acquired signal.

I whipped up a quick LTspice simulation to show what I mean. The whole scope system (including any cables) is represented as a simple single pole RC filter. The circuit on the left has a 10n pulse duration, the second on the right 1n, everything else is the same. Edge rates are in the same ball park as the AN-47 pulser (300ps rise time, more than an order of magnitude faster than the rise time of the RC circuit), but shape is not the same. My estimate for 10-90 rise time of the left circuit is 4.3ns, and for the circuit on the right is 1.1ns. Did the RC filter suddenly get that much more bandwidth? Is 320MHz bandwidth realistic for a 100ohm/20pF low-pass filter? The calculated bandwidth according to the 10ns pulse is 81MHz, close to the 80MHz calculated for the RC filter.

Solution: increase the pulse duration so it's much longer than the rise time. This ensures that the rising edge is over before the falling edge appears at the input. A charge line can be used to increase this pulse duration, see Janne's post:

Pulse is too short for narrow bandwidth scope. Suitable piece of coax can be used to lengthen the output pulse without affecting the pulse amplitude or rise time, see my measurements here.

[torch]

Hello everyone
I just statred +to nod my DS1052eE from version 02.05. SP2 and I have hit a wall. I have installed the 02.02. SP2 but can't connect to the scope with UltraScope. When I try to connect it asks me to choose from  ASRL1-INSTR or ASRL10-INSTR but either one doesn't work. I get the error "Can't Read from Device" and when I click ok I get "Device not found" I have uninstalled ULtrascope that came with cd and uninstalled "Visa462runtime.exe" reinstalled them and get the same errors. I did not get a request to install drivers.

What operating system are you running?

[flyingfrancisco]

Hi
I am running windows XP
flyingfrancisco

[torch]

Solution: increase the pulse duration so it's much longer than the rise time. This ensures that the rising edge is over before the falling edge appears at the input. A charge line can be used to increase this pulse duration, see

Unfortunately, he doesn't actually give his measurements in that post that I can see, which may be important since I'm not sure he's using the same stuff. The app note specifies 40" of 50 ohm "hard line or at least teflon-based &  rated for HF transmission", but he says he used RG-174/u, which is polyethylene based. (I think "hard line" refers to the thick direct-bury co-ax with the solid shield?) RG-178/u is teflon based though -- maybe it was a typo? Or maybe it's not actually that important?

What is actually more important: will this technique work with the AN47 circuit or does it require the AN94 version? There are some differences I assume the most important is the change in the cap value to 10pf and the addition of the damping resistor? The array of 200 ohm resistors in parallel add up to equal the 50 ohm termination of the AN47, why would he splay them out like that? Spreading the current distribution to minimize RF or something?  

Sorry for all the questions, but I'm an old dog trying to learn some new tricks. You have to speak s-l-o-w-l-y  

[torch]

Hi
I am running windows XP
flyingfrancisco

32 bit? 64 bit? SP1, SP2, SP3?

XP had some USB driver problems over the years. If you have 32 bit and followed the service pack upgrades in sequence, then those should be dealt with, but I seem to recall an issue if you skipped SP2 and went straight from SP1 to SP3. Not sure if it would affect this or not. 64 bit was pretty hopeless for drivers. Another potential issue is the physical location  of the USB port -- some devices don't like hubs and on some older motherboards there is actually only one port and the other connectors are downstream of a hub. Sometimes switching outlets helps -- look for one that is actually on the rear of the motherboard, rather than one mounted to the case and connected to the motherboard by a cable.

Do you have access to another computer you could try this on?

[flyingfrancisco]

Hi Torch
Do you think that it might be related to the fact that my Ultrascope software was fairly recentversion 2.08 I believe and that by downgrading my version fron 02.05 SP2 to 02.02.SP2 the Ultrascope software doesn't work with this version?

[torch]

Anything is possible! Mine is 00.01.07, which is available http://rapidshare.com/files/391131901/Ultrascope_for_DS1000E_Series.exe (as per the first post of this thread) You could try that and see if it makes a difference. Maybe Rigol modded the dll files to avoid the downgraded scopes?