Tektronix TWD120 100MHz dual channel digitizer teardown - opinions?

[GnatGoSplat]

This is a Tektronix TWD120 100MHz dual channel digitizer.  It's a PC-based oscilloscope from around 1994 with a SCSI-2 interface.  The software is 16-bit originally for Windows 3.1, but it'll run in any 32-bit Windows.  Only problem is it seems to crash when the window is resized beyond SXGA resolution (1280x1024).

MC68331 16MHz 32-bit Freescale MCU
4x M5M51008 (1Mbx8) 4MB SRAM
DS1630AB 32KB NVRAM
2x AM28F020 (256Kbx8) 512KB Flash ROM
NCR 53C90A SCSI Controller
MC68230 Parallel Interface/Timer
MC68681 Dual UART
8x IDT7164S20Y (8Kbx8) 64KB SRAM buffer for ADC?

My question: Is this good enough for a beginner to learn?  Or is it too outdated?  This is what I was told by another TWD120 owner.

The scope is better then I was thinking only a bit slower when you use more then 500 points like 5000.
I was very happy with my analog scope but this is not a bad piece of equipment I have to say it is far better then a 125mhz lecroy but those are from 1984 or so but claims to be 125mhz those have already a bit trouble displaying a 4Mhz square this one
According to the specs of my signal generator it should have a risetime in squarewave mode of less then 20ns and the scope says 10ns that is perfect, only good scopes display fast rise time like this.
A lot of features and it works up to 200mhz or more.

First I was shocked about the 20milliseconds fastest sweep time in the twd 120 viewer program because then it would be more bad then a homemade scope anno 1986 but now in the real program you have to use it goes down to 5ns that is not bad as you imagine you can zoom into the signal when you look at a 100Mhz signal, also the screen is big and clear compared to a small analog scope screen otherwise 5ns would be a bit low maybe.

I don't know enough about scopes to know if that makes it good enough to keep, or if I should invest in a lower-end Rigol.
I bought this a long time ago and have mainly just used its AUTOSET function in the software to view waveforms, but I'd like to learn more advanced scope usage.

[FenderBender]

Depends on the software? Is it unbearable? Okay? Good?

[GnatGoSplat]

The software seems okay, I always get the urge to maximize it which crashes it, but if I can resist the urge, it works fine.  However, so far I've mainly just used the AUTOSET button and changed volts and time scales.  I'm mainly just curious if the hardware sounds adequate, mainly from the part I quoted.  For example, he mentions a 5ns sweep time.  Is that adequate by today's standards?  Also, the official spec is 100MS/s.  Is that the sample rate in Mega-Samples per second?  If so, then a Rigol 1052E has a 10x faster sample rate.  That makes 100MS/s sound really slow, but is it adequate in real life?

[free_electron]

Some of the guts look like they have been recycled from tds5xx series scopes.

[T4P]

The software seems okay, I always get the urge to maximize it which crashes it, but if I can resist the urge, it works fine.  However, so far I've mainly just used the AUTOSET button and changed volts and time scales.  I'm mainly just curious if the hardware sounds adequate, mainly from the part I quoted.  For example, he mentions a 5ns sweep time.  Is that adequate by today's standards?  Also, the official spec is 100MS/s.  Is that the sample rate in Mega-Samples per second?  If so, then a Rigol 1052E has a 100x faster sample rate.  That makes 100MS/s sound really slow, but is it adequate in real life?

Not 100x ... 10x
anyway you are good for at most 20MHz ( yeah ... x5 is not any good )
USB scopes i have seen are mostly 200Megsamples anyway

[GnatGoSplat]

Not 100x ... 10x
anyway you are good for at most 20MHz ( yeah ... x5 is not any good )
USB scopes i have seen are mostly 200Megsamples anyway

D'oh!  Math fail, fixed it.
Are you basing that 20MHz on the 100MS/s or 5ns sweep?

[Fraser]

That unit looks typical Tektronix high quality kit 

I would love to own it myself. People appear to get very hung up on maximum sample rates and bandwidths when it is not always an issue. I am an RF Tech so prefer analogue oscilloscopes anyway, but I own several DSO's and Analogue/Digital CRO's. Each has its weak and strong points. The real time single sweep capability of your unit is around 5 to 10MHz with a 100 MS/s sample rate. That is not terrible, its just a little slow for todays faster clocking digital equipments. For general hobby electronics it is probably a good storage scope for lower frequencies. You could buy a half decent used analogue 100MHz Cathode Ray Oscilloscope to partner it and provide the higher frequency capability if needed. I personally find DSO's like my Rigol great for digital work but not so great for my RF work. I have used lower sampling rate DSO's for IF frequency work at 455kHz and also for battery discharge monitoring. They can also make great transient recorders. Attach a thermocouple adapter and you have a temperature graphing capability as well. Slow sampling rate DSO's are not without their uses 

Decide what sort of frequencies you want to look at, and whether storage is essential. Then you will know if you need to upgrade. I would suggest if an upgrade is the way forward, sell your Tektronix on ebay and you will likely be pleasantly surprised at the return you get on it. Then go buy the DSO or analogue scope of your choice.

[GnatGoSplat]

Thanks, I'll keep that in mind.  Though I probably have a lot to learn before I can even recognize whether I'm coming up on its limitations.

[T4P]

Not 100x ... 10x
anyway you are good for at most 20MHz ( yeah ... x5 is not any good )
USB scopes i have seen are mostly 200Megsamples anyway

D'oh!  Math fail, fixed it.
Are you basing that 20MHz on the 100MS/s or 5ns sweep?

100Megsamples ... single sweep at best 5-10MHz
20MHz is the maximum you can go before your waveform goes completely, maybe.
Due to Nyquist–Shannon sampling theorem it's at best 10-20MHz

[nctnico]

Actually you'll get aliasing artifacts. If you feed 101 MHz into a 100Ms/s scope you'll see a 1MHz sine wave. Usually this isn't a real problem though. IMHO its a decent piece of hardware. If you add an old Pentium 2 laptop and use a compact flash card as a hard drive you'll have a neat portable kit.

[joat]

I was wondering if anyone has a pdf manual for the twd120 they would be willing to share. I cannot find one anywhere. Thank you.

[rsjsouza]

Please apologize for resurrecting such old post, but perhaps others that were involved at the time will be notified by it.

About a month ago I received a unit in pristine condition with all manuals and software and was trying to make it work here. After lots of mucking around with some ancient systems I have here, I was finally able to put together an ancient Athlon 64 3500 with 1GB RAM, Adaptec AHA2940U and running Windows XP.

I followed the excellent instructions shown at the link below but with a minor caveat on XP: when installing Adaptec's WINASPI driver, you must also install the Windows 98 version to properly add the 16-bit version of this driver. Running install.bat X86 will do.

http://www.chichak.ca/file/Tek_TWD120.html

The Adaptec drivers above have a utility called aspichk.exe that works very well in detecting which versions are installed. In order for the unit to work all drivers must be installed.

On this configuration I didn't need to muck around with the shim driver mentioned in the reference above - it worked fine given the SCSI adapter on Windows XP was already on bus 0.

The unit with its User's manual (to be scanned when time permits):


With a FY6600:


Its software. It is pretty well done and the only perceived crash was trying to resize to full screen (SXGA 1280x1024).


Anyways, I hope this information helps recover many of these interesting units in the field.

[rsjsouza]

I finished scanning and uploading the TWD 120 Digitizer Instruction Manual 070-8821-00 to ko4bb at:
http://www.ko4bb.com/getsimple/index.php?id=manuals&dir=Tektronix/Tektronix_Miscelaneous