Introduction to the 5000-Series Oscilloscopes

From TekWiki
Revision as of 11:14, 11 October 2021 by Kurt (talk | contribs)
Jump to navigation Jump to search
Tektronix 5440, a later 5000-series model

The Tektronix 5000-series (5K) was introduced in 1971 as a series of lower-cost laboratory oscilloscopes with versatile plug-ins. They generally have larger display screens than the more expensive and higher performance 7000-series, but have very low bandwidth in comparison − 51xx mainframes are limited to 2 MHz, 54xx series offer 60 MHz.

The Basics of the 5000-Series

5000-series mainframes consist of an acquisition unit (the 5103 or 5403) containing the plug-in compartments and low-voltage power supply, and a display unit (e.g., the D10, D11, D12, D13, etc). This design differs from most other Tek equipment of the era except the 7704A from the 7000 series.

The idea was that the display unit could be sold separately for custom integrated uses by other manufacturers at a lower cost than the 600-series monitors, and at the same time provide a larger display area.

Within the Tektronix product line, the D10 and D11 display units (non-storage and bi-stable storage, respectively) are also used in constructing the 577 semiconductor curve tracer,

Nomenclature

Within the 5000-series, however, this design led to documentation problems as each oscilloscope ended up with two model numbers and two serial numbers. In 1977, Tek began to eliminate the serial number from the display unit for the scopes that were listed as “5103/D13” for example, using the serial number from the display unit for the complete oscilloscope.

Along with this change, Tek began referring to the various models with the acquisition unit prefix followed by the “D” number, such as 5110 for a 5103/D10, 5113 for a 5103/D13 or 5441 for a 5403/D41. Thus, documentation became much simpler.

Tektronix Service Centers began “grandfathering” this designation with the older models that had serial numbers on the acquisition unit, referring to them as the combined number (e.g., 5112) with the display's serial number, ignoring the serial number of the acquisition unit.

Unfortunately, all of this causes confusion today as service and operator's manuals are available for the 5103, D10, and 5110. In all cases, if you have the 5110 manual, you have all the operation and service information you need for that model. It will be pointless and take a lot more time, energy and money to find separate manuals for a 5103 and a D13 than it will be to find a 5113 manual.

Mainframe Numbering Convention

While the 7000-series has a certain convention for numbering the mainframes, the 5000-series has very little convention. The 51- and 54- prefixes indicate the respective series. After that, there is no numbering convention - unlike the 7000s, all 5xxx mainframes have three compartments, which eliminates any special meaning that would be attributed to the last digit of the model number.

The -10, -12, -13, -14, -15 and -16 suffixes have no special meaning at all. Some models are non-storage, some are bi-stable storage, some are dual-beam, some are both dual-beam and storage:

  • 5110 Single beam, non-storage
  • 5111 Single beam, split-screen bi-stable storage; 5111A model has updated circuitry, 50 Div/µs writing rate
  • 5112 Dual beam, non-storage
  • 5113 Dual beam, split-screen bi-stable storage, 20 Div/µs writing rate
  • 5114 Aha! Tektronix never made a 5114! Who knows why they skipped over it to the 5115?
  • 5115 Single beam, split-screen, bi-stable storage (>200 Div/µs writing rate)
  • 5116 Single beam, tri-color display using an LCD shutter
  • 5223 Single beam, digital storage
  • 5440 Single beam, non-storage
  • 5441 Single beam, variable-persistence storage
  • 5444 Full dual beam (independent horizontal deflection systems using the 5B44 timebase)

(Option 3 enhances the writing rates to 200 Div/µs on most 5100-series storage mainframes.)

Plug-In Numbering Convention

5000-series plug-in numbering coincides more with that of the 7000-series. The first digit, a “5”, designates the plug-in as being for use in a 5000-series instrument. Unfortunately, there is no standard indication of whether or not it is compatible with the 5400-series. The second digit marks the general use of the plug-in. “A” units are preamplifiers; “B” units are timebases; “CT” designates curve tracers; “D” marks digital instruments; “L” is used to denote spectrum analyzers; and “S” denotes sampling plug-ins. The last two digits, however, depart from any standardization. It would have been nice for preamps to have a 1, 2 or 4 to denote number of channels for the third digit and then use the last digit to denote function, such as differential. But the last two digits are meaningless other than to differentiate one plug-in from another. The timebases are the same anyway, so there is no point of any further discussion.

The Two Faces of 5K and Compatibility

The 5000-series is divided into two product lines, the low-frequency 5100-series and the high-frequency 5400-series. The 5100-series has a maximum mainframe bandwidth of 2 MHz and lower bandwidths may be inflicted with plug-ins such as the 5A22N. The 5400-series has a maximum mainframe bandwidth of 60 MHz, again this figure is at the mercy of lower-bandwidth plug-ins.

The lower-bandwidth plug-ins designed for the 5100-series (commonly referred to as “5100-series plug-ins) are the preamplifiers with suffix numbers of “2x” and lower, such as the 5A22N, 5A15N or 5A13N. Timebases for this series will have suffixes of “1x” and lower, such as 5B10N or 5B12N. In addition, the 5CT1N semiconductor curve tracer, 5L4N spectrum analyzer and 5S14N sampling unit are available for use in this series.

Four plug-ins of the 5000-series directly mirror “twins” in the 7000-series: the 5CT1N (7CT1N) curve tracer, 5A22N (7A22) differential amplifier, 5A13N (7A13) differential comparator, and the 5S14 (7S14) sampler. Front panel layout and function of these “twins” are nearly identical, although the bandwidth of the 7A13 is significantly higher than its 5K counterpart. Note also that the 7000 plug-in interface is significantly different from the 5000 series, so the plug-ins are not identical but may use common parts together with adapters, like in the 7S14 that contains the full 5S14 board.

Nomenclature

“N” at the end of any Tektronix model number indicates a unit that does not have on-screen readout capability. All of the low-bandwidth plug-ins carry that designation, but for some reason, none of the 5100-series mainframes ever have the “N” designation even though they have no readout capability. All 7000-series products without readout capability have this designation on everything: 7403N, 7603N, 7A18N, 7B53AN, etc. It is not known why the 5110, for instance, is not the 5110N. One must suppose that it's simply because none of the mainframes ever have or can have readout capability. But this doesn't explain why the low-bandwidth plug-ins, none of which have or can ever have readout do have the “N” designation, with the exception of the 5A21 and 5A22 which have a readout option for use in the 5400-series mainframes.

Compatibility

5100-series plug-ins may be used in any (5100- or 5400-series) mainframe in the 5K series. They will severely limit the bandwidth of the 5400-series, but they offer capabilities to the 5400-series that the series would otherwise not have, considering the function of the 5A13N, 5A22N, 5CT1N, 5S14N and 5L4N.

The 5400-series has its own small set of plug-ins dedicated only to that series. The 5400-series plug-ins absolutely cannot be used in the 5100-series and in fact, have interface connector keyways plugged to prevent them from being seated into a 5100-series compartment. These keyways must never be removed or altered.

The 5400-series plug-ins are those that have suffixes of “3x” and larger for the preamps and “4x” and larger for the timebases. There are no other standard plug-ins that are not preamps or timebases for the 5400-series with the exception of the 067-0680-00 mainframe normalizer calibration fixture.

At this point, one may be curious and wonder why the 5400-series plug-ins cannot be used in the 5100-series mainframes. A close look at the interface connector on both series will show that the 5400-series has two extra power supply voltages, ±15 V on interface connector pins 6A and 6B. 5400-series plug-ins will not find those two supply voltages in a 5100-series mainframe, so they will not operate correctly. Those two pins are otherwise unconnected in a 5100 scope and whether or not one could modify one by building a bipolar 15-volt supply for those two pins and actually get a 5400-series plug-in to work remains an unanswered but rather pointless question.

TM500-Series/5000-Series Compatibility

The mechanical aspects of the TM500 line of modular instrumentation and those of the 5000-series are identical except for the interface connector keying, however, TM500 instruments and 5000-series instruments are electrically totally incompatible with each other, and any attempt to operate one type of plug-in in the other type of mainframe may result in the destruction of the plug-in, mainframe or both.

(There has been an instance of a TM500 frequency counter installed in a 5000-series mainframe with the LED display bright and presumably active, as the photo illustrated for some items being sold on an on-line auction, but it appears likely that this image was “photoshopped” by the unscrupulous seller.)

In the 5000-series, all power supply voltages are developed by the mainframe and supplied to the plug-ins as regulated DC voltages of +5 V, ±15 V (5400-series) and ±30 V. There are no power supply circuits other than supply decoupling and small zener regulators in the plug-ins.

On the other hand, the TM500-series power modules (the “mainframes”) supply AC and raw DC voltages to the compartments and no regulated voltages, and these voltages are not supplied on similar interface connector pins as the 5000-series. All of the power supply regulation control circuitry is inside the individual plug-in for customized voltages, since each plug-in instrument is a radical design departure from the other instruments. The power module also contains one NPN and one PNP power transistor for each compartment for use by the instrument installed as a power supply regulation component, a design to keep the heat out of the plug-in and in the power module where it can be properly dissipated. None of this is even similar with the 5000-series.

However, this writer, plagued with intense curiosity, wondered if a 5CT1N could be modified to operate in a TM500 power module as an independent curve tracer with an external display. After a lot of study and design work, it was discovered that such as thing could be done. But it took modification of nearly every interface connector pad, cutting scores of circuit board runs, adding scores of jumpers and designing a “piggy-back” circuit board with extra power supply, warning indicator and signal interface circuitry. It was done, the project was a success and it was self-designated the “CT501” a rogue, non-Tektronix product. In the near future, you should be able to find the documentation for this radical conversion project on the pages of Tekwiki. It is a very intensive and laborious project requiring a lot of attention to detail and a lot of patience, and there's still no guarantee that it will work because of differences in production runs of the 5CT1N circuit board. But this is all an attempt at an illustration to show how the two systems are not the least bit compatible.

Plug-Ins and Bandwidth

The 5100-series mainframes are limited to a bandwidth of 2 MHz. As low as that bandwidth is relative to the rest of the Tektronix product line (there have been probably less than a dozen scopes with lower bandwidth over the life of Tektronix), many of the plug-ins lower the system bandwidth even more! Full bandwidth of a 5100-series mainframe can be had only with the 5A15N, 5A18N and 5A19N. The other preamplifiers limit the system bandwidth to 1 MHz, even less if you use a 5A22N and selectively limit the upper frequency cutoff.

Some may think that this low bandwidth makes the 5100-series worthless. In reality, there are five things that make the 5100-series attractive:

  1. Those who work in the low-frequency arena of audio sound systems, LF and VLF radio, mechanical transducers, hearing aids, etc. will find the lack of HF noise in the 5100-series to be a blessing;
  2. The plug-in versatility of the 5100-series can be matched only by that of the 7000-series, and then at a far higher price in cost and complexity;
  3. A large display screen, internal graticule and fine, sharp trace;
  4. Simple and easy troubleshooting and repair, usually involving no custom, complex and hard-to-find Tektronix-made custom ICs;
  5. Three nice plug-ins, the 5CT1N semiconductor curve tracer, the 5L4N audio spectrum analyzer and the 5S14N sampling system give the 5100-series power that you can't find in any of Tek's portable scopes. The 5S14N increases the effective bandwidth of the system to around 1 GHz.

Some Interesting Notes

The dual-beam 5444 was offered for sale for just one catalog year at $3,330 and then was gone! The 5B44 was a plug-in developed for the 5444 to allow access to BOTH sets of horizontal deflection plates for full dual beam operation similar to the 7844 or 566. This plug-in appeared in the same catalog as the 5444 (for $895) and stayed for a full five years, even though the 5444 was the only scope in which it would work correctly. By the time it was deleted from the catalog, the price had risen to $1,020. It would be a good story to hear.

The 5CT1N semiconductor curve tracer was introduced in the 1972 catalog and the price slowly rose from $350 to $375 over four years. Then in 1976, the price shot from $375 to $600! The next catalog in 1977 increased it to $630 and then a year later in 1977, it dropped down to $510!

The cheapest Tektronix instrument on record is probably the 5A24N, a spartan preamplifier that originally sold for $25 in 1971! Ten years later, the price had risen to $160.

History of the 5000 and subsequent series of instruments by Ahne Oosterhof

It was late 1969 when the low frequency scope group was about finished with the 5030 scopes (1969), when Val Garuts and I prepared the product proposal for what first became to be called the LF-Scopes, then the D10/5103N and finally the 5000-series. The proposal went two or three times to the Engineering Council before it was accepted (Jan 1970). The group was formed under leadership of Jerry Shannon. I had built a cardboard model of what we envisioned, half the scope for the CRT, etc. and the other half for three plugins and power-supply, all within 5.25 inch height, a standard height increment for 19” racks, and the mainframe with three plugins for less than $1000. Above the plugins a strip of space was envisioned for plugin on/off switching, etc. I remember Larry Weis walking by my bench suggesting that we should make the plug-ins maximum height and build that switching functionality into the individual plugins. Seemed like a good idea. One year later (Jan 1971) with about 12 people on the project (EE, ME, manual, ecb) we had two scopes and three plug-ins in production using several new manufacturing methods and many new low-cost components. Almost by default I was working as the lead person on the mainframes and Gary Vance performed that function for the plug-ins.

Throughout that project some big hurdles had to be overcome, e.g., using tin plated rather than gold plated circuit boards, soldering active components in the board rather than using sockets, using low cost open potentiometers rather than fully closed ones. I had asked the mechanical designers under Bud Deibele to make the frame weak enough that it would fail any shake, shock and transportation test. Then make it a little stronger to barely pass those tests. The pieces of the half frames were riveted together rather than bolted and to ship an empty frame required a cardboard insert to prevent distortion of the frame in shipment. All in the name of lowering costs.

In the assembly area kits for 10 plug-ins were issued to individual assemblers and after finishing these, the assemblers tested those themselves which led to fewer assembly errors: smoke or smells signal loudly that you have made a mistake! Even though it took some effort to convince the manufacturing manager, Al Peecher, that it would work and he made it a go.

During the design phases Jerry suggested that we make many presentations throughout the company to Area Representative groups. That paid off big, because as people learned about our project and our challenges they were more than willing to provide us with any help we needed.

As we reached the finish of the initial project we looked for product ideas for the future and that kind of planning always was done as a group effort. It led to dual beam scopes, additional storage scopes and various plug-ins, etc. to extend the 5000 series.

We also had noticed that some customers bought display sections only, to build those into their instrument systems. That became somewhat of a headache, so we designed a power supply for the displays and turned that idea into a new line of monitors, an extension of old the 600 series. That replaced the dormant 601 and 602 displays. Soon after that we split the group into two business units, scopes and displays.

We also considered a ‘plug-in warmer’ (like in the 530/540 series) but that was not really needed as the plugins were completely transistorized and therefore immediately ready to be used. But it made us consider additional possibilities like making a multi-meter and other measurement tools and signal source plug-ins. Soon this turned into a lot more and became the TM500 series (1972?), using the same design and construction methods as the 5000 series and this line became another separate business unit.

Each of these product lines were successful and stayed in the catalog for about 20 years.

Throughout all this Jerry Shannon encouraged a free flow of ideas; nothing was out of bounds to be discussed and tried. As an example, Matt Zimmerman in his spare times put together a transistor test plugin, which became the 5CT1. Throughout Jerry liked to take an opposing view of whatever anybody said. He was willing and able to take any position on any subject from design to politics and if you thought things were blue he would try to convince you they were red. I expect he did this to make sure we were strong in our positions and opinions. In doing so he taught us a lot about dealing with and managing people. He made it clear to us that we were as design engineers and project leaders standing in for the owners of the company, as we were making decisions, however small, that impacted the health of the Corporation. He also made it clear he expected us to make the right decision more than 80% of the time, but did not expect 100%, as it would take too long to get there. Risk was allowed.

Yes, he guided us, but in a very unobtrusive way. He let us find our own way, but was ready to clear a path when needed. In my life I can think of only three people who I consider my mentors: Professor Oostdijk from my college years, teaching (forcing?) me to study, Professor Dwain Richins from the “Tek University”, teaching me a lot about communicating clearly and in a precise way in the English language and Jerry Shannon. He taught me a lot about managing people and taking care of unexpected situations.

In my second career, standing alone as entrepreneur/owner/general manager of a company (A-Laser, Inc.) I have made a lot of use of their teachings.

A description of most of the 5000 series products with a timeline can be found in this website: http://i9t.net/5000_scopes.html