Old Logic Analyser Help

[MattHollands]

Hey!

I recently acquired an old HP/Agilent 1662AS logic analyser. I'm moving back to the UK in summer, and it'd be nice to take it home, but it's so big! So it got me wondering about replacing the CRT with an LCD. Just by sticking an FPGA in the middle. So I opened it up, and now I have some questions. Attached are some pictures to help illustrate my questions:

So it appears to consist of 3 boards. 2 which seem to do processing, and a third (on the right) which seems only to drive the CRT.

The CRT board has a big black thing on it with a large wire connecting it to the CRT via a suction cup. What is this thing? I assume it generates a very high voltage for the CRT.
Apart from connecting to the CRT, the CRT board seems to connect to the other boards by only one connector (the grey shown in the images). That implies that this connector carries video signals and power, and that the power is enough to drive a CRT, is that likely? I would've imagined they would've had larger power supply wires.

Thanks for any help!

[timb]

The large wire attached to the monitor with a "suction cup" carries the primary anode voltage, usually 8 to 25kV. It's not actually a suction cup, but some type of rubber potting that covers the electrode (which is attached to the aquadag).

These are magnetic deflection type CRTs, hence the yoke around the neck of the tube. The good news about that is the screen most likely uses some sort of simple arrangement to produce video: H-Sync, V-Sync and a video pulse (Z-axis) to turn the beam off and on.

I have an old Sony/Tektronix 308 logic analyzer here that works just that way. In fact, the CRT driver board is completely self contained with HV supply and everything. The CPU board simply sends the H/V-syncs and Z-axis signal. If yours is similar, it would be easy to remove the CRT and driver board and pick those three signals off with an FPGA or even MCU (if the resolution is low enough) to feed an LCD.

Do you have a copy of the service manual? Start there.


Sent from my Tablet

[nctnico]

If you can find the digital signals which drive the DACs for the CRT signals then you can attach a CPLD to them and drive an LCD from there. I have done several of these conversions and the difficulty depends on resolution and whether or not the display is interlaced. However a CPLD like the Xilinx 9500XL series is good enough and since it is 5V tolerant it is ideal to interface to this kind of old style logic.

[MattHollands]

I haven't managed to get a hold of the service manual, but I have managed to for the 1650B model (this is 1662AS) and it looks very similar.

http://www.doe.carleton.ca/~nagui/labequip/la/1650B_Service%20Manual.pdf

If you go to page 102 it describes a 20 pin cable with certain inputs, which looks identical to the one on my model. I am pretty confident my model will carry the same pin outs but I want to check.

Is there any danger in running the model without the outside case on? It says "Xrays generated in this device are sufficiently screened" on the case. Would this continue to be true if I ran it without the outer case on? Of course there would be dangers about touching high voltages.

Could I connect the probes of the logic analyser to the cable so that I could use it to see what's going on inside the device? Or is there some problem with that?

[MattHollands]

I have done several of these conversions and the difficulty depends on resolution

Looking at the spec, it looks like it has hsynce, vsync and two wires indicating if the pixel should be dark, medium or light.  Do you have any suggestions for replacement LCD panels?

[vk6zgo]

I haven't managed to get a hold of the service manual, but I have managed to for the 1650B model (this is 1662AS) and it looks very similar.

http://www.doe.carleton.ca/~nagui/labequip/la/1650B_Service%20Manual.pdf

If you go to page 102 it describes a 20 pin cable with certain inputs, which looks identical to the one on my model. I am pretty confident my model will carry the same pin outs but I want to check.

Is there any danger in running the model without the outside case on? It says "Xrays generated in this device are sufficiently screened" on the case. Would this continue to be true if I ran it without the outer case on? Of course there would be dangers about touching high voltages.

Could I connect the probes of the logic analyser to the cable so that I could use it to see what's going on inside the device? Or is there some problem with that?

The "X -Ray " thing is to cover HP's backside ----that part of the LA is really just a CRT-type Monitor,like you used to have sitting on your Computer desk,most of which were in a plastic cabinet!.

In turn,a Monitor is effectively a TV set with some of the "guts" missing.
TV Servicemen spent hours around such things outside the cabinet.

Receiver type CRTs like this will produce minimal levels of X-rays,& utilise quite thick glass with some lead content in the front part of the tube,where any such rays would be generated.

[MattHollands]

OK cool! Thanks

I've been looking at CPLDs to develop this on. I would like to find cheapish dev board (not in the 100$ range like FPGA boards).

The best thing I've found so far is http://www.seeedstudio.com/depot/XC9572XL-CPLD-development-board-v1b-p-799.html which has 72 macrocells, however from reading around, this might not be enough.

Does anyone know of any nice 144/200+ macro cell cpld dev boards?

[poot36]

Try looking for the HP 1660A service manual.  The 1650 series are vastly different.  I have a 1661C and its video connector has 12V, Ground, H-sync, V-sync, and I think 0.7V p.p analog video.  The later 1660E series supported color, mine is 8 shades of grey.  If I remember from the service manual the screen is interlaced and runs at 15khz horizontal.  If yours is really a AS model you have a scope in there as well.  You can use it to verify the signals coming out of the video connector.  I think that it would be fairly easy to convert the signal to composite video and feed that into a cheap 7" or so lcd video monitor.  The resolution is just above 500 x 300 if I recall.

[MattHollands]

Good call! Turns out I was searching too specifically for the 1662AS datasheet as opposed to the series. Thanks Turns out its a 500x240 display with hsync, vsync and video. It is also 8 shades of grey. I can't see anywhere telling me if it is interlaced or not, or how fast it runs, but I can try and figure that out from the video signal.

[BloodyCactus]

the 167x series already has LCD, look for that manual, maybe those changes can be backported..

or might be cheaper to just get say 1670g for 350$

[poot36]

I was reading the theory of operation and came across this blurb "CRT Controller and Display RAM
A Brooktree BT475 RAMDAC (datasheet here: http://www.datasheetarchive.com/dlmain/Datasheets-8/DSA-151625.pdf) color palette and a National Semiconductor LM1882CM video
frame generator (datasheet here: http://pdf1.alldatasheet.com/datasheet-pdf/view/8687/NSC/LM1882CM.html) control the CRT. One of the RGB outputs of the color palette provides the
eight-shade gray scale display. The video frame generator provides the horizontal and vertical
synchronization timing signals.
The display RAM is a 256Kx8 video RAM and stores all of the pixel information used by the
color palette. A serial address counter and an address multiplexer control the DRAM
addressing. At the conclusion of each video frame the vertical sync signal from the video
generator resets the serial address counter and a new frame is generated."  and this one "CRT Monitor Assembly
The CRT Monitor Assembly consists of a monochrome CRT and a monitor driver board. The
monitor driver board provides the biasing and control signals for the CRT. Pixel information is
stored in the display RAM on the CPU board and is routed to the monitor driver board through
the acquisition board and the sweep cable."  Could this be moded to provide color output?