开云体育

I'm looking get get set up to be able to calibrate my own scopes. In reading the 485 manual a high frequency constant amplitude signal generator 100-350mhz is needed. If in the future I pick up a faster scope Should I get something more than 350mhz? Any recommendations? Asking here will save me from buying something that wont work for me. Thanks guys.

On 2018-03-02 3:31 PM, Ed Breya via Groups.Io wrote:

I'd recommend just using almost any analog scope that has X-Y capability for display. For low frequencies including audio, the response of the V and H should be virtually identical - you can easily check to be sure. Scaling is no big deal, and can be set up with the scope's range, variable, and position controls to whatever is needed. Even if not, a simple interface circuit should do the trick. You can also use a DSO if it has X-Y, to record very slow or single-shot experiments. Or, an analog storage scope.

I've found X-Y mode on a digital scope (TDS460A at least) to be
extremely unsatisfying. It seems to be far from the sweet spot of what
DSO's can do.

I've always had a soft spot for analog computers. Years ago, I had planned to build one, with lots of the usual function elements, and also some digital (kind of cheating) enhancements like A/D-EPROM-D/A operators for non-linear and arbitrary functions. The problem was that there were so many possible options that my tendency to overdo things would have made it a mind-boggling, complicated mess of functions and interconnect issues. So I never did it, but it's fun to think about.

When I think about such a project I think that a useful role for digital
would be in the (re)configuration of a module fabric (e.g., analog
crossbar switches).

There is also this:

But this is probably off topic by now...

--Toby

Until this discussion came up, I had no idea that analog computers are still so popular in some circles - looking online a little now, I see that they are alive and well to some extent, even though PCs and SW can do all that and more. There's something to be said for the real thing, hands-on experience.

BTW, I stumbled upon a manual for the GP-6, that appears to include all the circuit schematics. If you haven't found this already, here it is:


Good luck,
Ed

On Fri, 02 Mar 2018 18:39:11 +0000, you wrote:

3D printing can be a bit picky about settings. The smaller the layer
setting the denser the part, and the more rugged it will be. Extruder
temperature also plays a part.

Harvey

On Thu, 1 Mar 2018 at 15:59 Malcolm Hunter <malcolm.r.hunter@...>
wrote:

It can be done. I've 3D printed feet for the 465 based on this model I
"designed" based on the original feet:
. Printed in PLA, they are
functional but I wouldn't bank on them holding out for every day use.
Someone with a bit of 3D modelling ability and an original foot should be
able to craft a suitable replacement for the 335 too. Printed in something
a bit more durable would work and even if they weren't as good as the
original, you could have plenty of spares.

There are now services that'll demand-print and ship items, like e.g.
ShapeWays, which will print in a variety of materials. It'd be nice to have
a stash of 3D models for Tek replacement parts cataloged someplace, and/or
to have them listed on these print services.

I modeled the PCB bracket I needed for my FG504, but the PLA prints I got
from the 3D printer I have access to delaminated as soon as I put a screw
to them. The ShapeWays nylon print, however, worked just fine (see <
>),
and I know other people have had luck with the same print.

I priced out your thing on Shapeways, and the two materials/methods that
look likely are:

- Strong & Flexible Plastic Black (Selective Laser Sintering): $8.98.
- HP Nylon Plastic Black (Multi Jet Fusion): $17.96.

though perhaps it'd be possible to bring down the per-item cost for e.g. a
set of four by optimizing the design...

I'd recommend just using almost any analog scope that has X-Y capability for display. For low frequencies including audio, the response of the V and H should be virtually identical - you can easily check to be sure. Scaling is no big deal, and can be set up with the scope's range, variable, and position controls to whatever is needed. Even if not, a simple interface circuit should do the trick. You can also use a DSO if it has X-Y, to record very slow or single-shot experiments. Or, an analog storage scope.

I've always had a soft spot for analog computers. Years ago, I had planned to build one, with lots of the usual function elements, and also some digital (kind of cheating) enhancements like A/D-EPROM-D/A operators for non-linear and arbitrary functions. The problem was that there were so many possible options that my tendency to overdo things would have made it a mind-boggling, complicated mess of functions and interconnect issues. So I never did it, but it's fun to think about.

Until this discussion came up, I had no idea that analog computers are still so popular in some circles - looking online a little now, I see that they are alive and well to some extent, even though PCs and SW can do all that and more. There's something to be said for the real thing, hands-on experience.

BTW, I stumbled upon a manual for the GP-6, that appears to include all the circuit schematics. If you haven't found this already, here it is:


Good luck,
Ed

Thanks for the extra info. I certainly wish I could peer over your shoulder, I'd probably learn a lot.

A very cheap X-Y option right now is a 1720 vectorscope, if that has the bandwidth you need; it's one of the devices I'm using to test my circuit with. They are small, lightweight, and currently cheap on ebay (there's a $10 one right now). The 60x family seem to be very hard to find these days (although a friend bought a 603 from Sphere recently, which I am about to try out).

--Toby

On Thu, 1 Mar 2018 at 15:59 Malcolm Hunter <malcolm.r.hunter@...>
wrote:

It can be done. I've 3D printed feet for the 465 based on this model I
"designed" based on the original feet:
. Printed in PLA, they are
functional but I wouldn't bank on them holding out for every day use.
Someone with a bit of 3D modelling ability and an original foot should be
able to craft a suitable replacement for the 335 too. Printed in something
a bit more durable would work and even if they weren't as good as the
original, you could have plenty of spares.

There are now services that'll demand-print and ship items, like e.g.
ShapeWays, which will print in a variety of materials. It'd be nice to have
a stash of 3D models for Tek replacement parts cataloged someplace, and/or
to have them listed on these print services.

I modeled the PCB bracket I needed for my FG504, but the PLA prints I got
from the 3D printer I have access to delaminated as soon as I put a screw
to them. The ShapeWays nylon print, however, worked just fine (see <
>),
and I know other people have had luck with the same print.

I priced out your thing on Shapeways, and the two materials/methods that
look likely are:

- Strong & Flexible Plastic Black (Selective Laser Sintering): $8.98.
- HP Nylon Plastic Black (Multi Jet Fusion): $17.96.

though perhaps it'd be possible to bring down the per-item cost for e.g. a
set of four by optimizing the design...

Toby,

Documentation is just a notebook, a few pics before disassembling anything important.
Analog computers are two Comdyna GP-6. FWIW, the system could modeled in MatLab, Mathematica, or even MathCad. This project is modeling some audio transfer functions - it will be quicker for me to optimize by seeing and hearing the residual errors. For the display a vintage Tektronix 503 will do; has identical X and Y amplifiers so additional scaling circuitry and their effects is eliminated.
Also checking out Rolynn's suggestion of a 602 or 604 monitor.

Thanks, Ron

On 2018-03-02 12:25 AM, Ronald Carlson wrote:

Anyone have a Tek 503 that's looking for a new home?

The 503's specs fit nicely as an output display for an analog computer project I'm working on.

I'm curious about your analog computer. Are you
documenting/blogging/video'ing this anywhere? What is the goal?

I'm also working on circuits to drive X-Y displays, but at a very basic
level since I'm an electronics beginner.


--Toby

Or, perhaps there's a better, newer model the group could recommend?

The identical low frequency response horiz and vert amplifiers with a
square display make it initially an interesting option.

TIA, Ron

It is made on laser engraver. I have a friend that work on this machine and I make a corel with outlines and what need to be on scale.
it is made from black PVC foil past on Plexiglass.
First cut the foil on low power to remove the foil only and after this cut outline on big power.
I'm not sure which one will illuminate better, but will test all 3 before to choose one. Think this one with black foil back will be better then other because will illuminate only the scale on top of the LED.
I hope in next week to find a time to go to make the knob and will make some pictures with all installed.

I have been refurbishing some late 70s-early 80s HP gear and I only replace bad capacitors per ESR or Q meter readings. A schematic helps in making sure there are no other elements in parallel that would skew the results. Many power rail caps are in parallel so I just lift their one end and measure that way.

I have found several bad tantalum slug capacitors - I replace those with modern equivalents that often have similar part numbers and look very similar or identical. Those are not particularly cheap and I always replace the hermetic ones with new hermetics, not the cheaper yellow plastic encapsulated ones.

If a given type of a capacitor has failed more than once in a given instrument, I replace all - if there’s less than 5 or 6 total. Otherwise I do a judgment call based on circuit types and the stress the part is likely to experience.

The failure mode for tantalums that I checked was almost always high ESR or high DF, and capacitance changes were secondary and not always present.

I also test all electrolytics after 24hrs of preforming/burn-in at room temp. They seem to fail randomly and I also shotgun those that see lots of stress if any compatriots have failed. For me, they almost always fail with both ESR/DF and capacitance changes. I would still advise having a good ESR meter in addition to capacitance meter.

In some units there is a lot of failed electrolytics, in some others - few if any. I don’t have enough experience to see a clear pattern there yet.

Kuba

The LO input to the mixer is also dead. It comes into the mixer through a cable/plug and it's still dead while unplugged. There are several test points on the LO circuit . The manual has a few reference wave forms I can check. In summary, both the LO is dead and the mixer's output transformer isn't even passing the basic AF wave by itself. But the transformer isn't open. I'll dive into these over the weekend and see what I can find.

Sounds good. I terminated it with 600 ohms and now it's perfect. Thanks for pointing me in the right direction.

Rick

Dennis' suggestion of using add and invert mode on the oscilloscope
works well depending on how much sensitivity you need.

If you have a 7000 mainframe, then a 7A13 combined with standard x1
probes may be acceptable. Standard x10 probes have the problem of not
having matched attenuation which spoils the common mode rejection.

Another alternative I have occasionally used is to build the probe
into the circuit using a fast instrumentation or difference amplifier.

Dave,

Thanks, yes a couple of them sold that would be just fine. Repair project is OK. I'm standing by and eventually I'll be providing one a new home and tlc. Either private party, hamfest, or the 'bay.

Ron

To second Dennis, this is exactly what we do, and we’re a professional audio archive lab.

George

Massdrop had one a week ago for $140.



I was actually quite surprised to see such a specialized piece of equipment
show up for sale there.



If it's not available now, be patient, it'll come up again.



Menahem

Dave-

You might consider going to ham radio swap meets. A really great FET differential probe is the Tek P6046. I picked up one for less than $50 if I remember correctly and it is perfect. Many people do not know what they are and that accounts for them selling so cheaply. It takes some time to find one but at least you'll be rewarded with a great tool once you do.

Phil

There were some selling very reasonably on eBay not too long ago; they
still might be on there. Have you checked?

Dave Casey (who is not currently selling any scopes on eBay)

On Thu, Mar 1, 2018 at 11:25 PM, Ronald Carlson <
ronald.carlson.54@...> wrote:

Anyone have a Tek 503 that's looking for a new home?

The 503's specs fit nicely as an output display for an analog computer
project I'm working on. Or, perhaps there's a better, newer model the
group could recommend? The identical low frequency response horiz and vert
amplifiers with a square display make it initially an interesting option.

TIA, Ron

Anyone have a Tek 503 that's looking for a new home?

The 503's specs fit nicely as an output display for an analog computer project I'm working on. Or, perhaps there's a better, newer model the group could recommend? The identical low frequency response horiz and vert amplifiers with a square display make it initially an interesting option.

TIA, Ron

There is no such thing as a cheap differential probe. They have to be matched across a wide frequency range to have good CMRR. If you want a cheap CMRR the solution may be on your scope. All dual trace scopes have an ADD function for both traces. They all haave a way to invert a trace on one of the channels. So switch to Add, Invert one channel, and use the gain knob of one channel to adjust the trace to a straight line when you put a signal in through two same number probes - any two will do as long as they are the same (two P6109 probes for example).

Voila, a poor man's Differential probe.

As long as there isn't too much of a voltage difference between the two channels the waveform will stay on the screen where this will work OK. It isn't much but that is probably the best you are going to get when you require "cheap"

Dennis Tillman W7PF