The next best-kept test equipment series

[suddenseer]

I don't even own a service monitor anymore, let alone "dedicated". You da man. I was a moocher. I always had a job at a facility with loads of lab equipment to work on my stuff. I even used my company's hot air soldering equipment, would hit a .032" dot without bridging. You could open up your own business with the hobby toys that I observed.

[N8YX]

Though it's made for mechanical measurement signal conditioning, the AF501 Bandpass Filter/Amplifier also functions nicely for general purpose AF work. Of course, you'll need two of them (along with two -0938 Calibration Fixtures, featured on the previous page) if you wish to perform simultaneous L/R channel measurements of a stereo power amplifier. An evaluation of the input amplifiers/conditioning circuitry - specifically, quantifying channel leakage and cross-mixing - would be a typical application. To do this you'll need two signal sources (SG505s or equivalent), the pair of audio notch/peak filters we're discussing here, a dual-channel 'scope and a pair of DM501As for RMS measurements. Throw in an AA501 for distortion measurements and you've eaten up two TM506 mainframes worth of space.

http://www.barrytech.com/tektronix/t.../tekaf501.html

tekaf501front.jpg

[KG4CGC]

MacGyver comes to you, for advice.

[N8YX]

MacGyver comes to you, for advice.

My advice:

Don't pay eBay prices for this stuff...especially from those test-equipment resellers who think the equipment is still worth a considerable fraction of its new price.

I just grabbed another one of the AF501 units for $125 and even then I think I overpaid for a "very rare" (hah!) unit...considering there are several listed at the moment. Naturally, as soon as I won this unit two more were listed...for higher opening figures still.

And there are a pair of TM504 power modules listed as I type this. In addition to the other modules included with each (all four slots filled) is an AF501. $300 B-I-N or Best Offer. This is a realistic price for the equipment being auctioned.

[N8YX]

Bump.

Going to put this one out there as a reference and tale of "how to fix".

The Tek PS-5004:

500x500.jpg

A '5004 is a very useful gadget to have when you're doing such things as designing or servicing comparator or threshold-detection circuits and must employ a voltage/current source with greater than 1mV/1mA resolution.

They're NOT worth the $200-500+ prices which are being asked, especially for an "as is" unit.

I took a chance on a $99/shipped offering which read "32.740" upon power-up. A normally functioning '5004 should read "0.000". when initialized.

Replacement of all electrolytic capacitors in the -25.5v, 12V, 8V and 5V internal supply circuitry was necessary. Once this was done, the supply was plugged into a TM5006 mainframe using extender cables and powered up. It worked until three transistors in the 12V control supply section decided to go up in smoke.

Replacing them fixed the smoke issue but the display read "32.740" afterwards.

Here's the tricky bit:

The device is set up with a current DAC/current comparator and a voltage DAC/voltage comparator operating in bridged mode, and the supply itself either operates in a constant voltage or constant current manner. If the 8V supply which powers these comparators fails, the output follower device is driven to the supply rail...hence the high voltage reading.

That 8V supply is an MC78L08ACG - TO-39 package. Replacing it with a TO-92 version ( 78L08 ) works just fine.

Moral of the story: If you get one of these units and find the power-up readout doing what I've described, check all supply voltages. Additionally, check the TM9914 GPIB interface IC on the Programmer Board. If it's defective it'll draw the 5V supply low and the front panel will cease to function properly.

[N8YX]

Finding myself with a bit of spare time over the past month, I dug a little deeper into the "to-do" pile. First off was replacing the display in a DC508 counter:

tekdc508front.jpg

The OEM readout uses three Siemens DL-883A 3-digit displays - which have long since become unobtanium. One of the guys on the Yahoo Tek500 Forum made a limited run, aftermarket readout board using discrete single-digit SMD LEDs which fits both the DC508 and DC508A counters.

I HIGHLY recommend this board if you have a '508 series counter. Installation took a couple hours, mostly due to the fact that my big fingers are a little challenged in small spaces - and there isn't a lot of room to work with in those smallish modules.

FWIW, if one can find a junker DC-503 (not the -'A' variant) and one has access to an ESD-safe vacuum desoldering station, the readout board may be removed and the center display can subsequently be desoldered and used to fix your '508.

As it stands, I now have a pair of readouts with which to fix my '503s should they ever go bad.

[N8YX]

Next up was a DC-5010 counter.

Any of the xx-5xxx series of plugins will only work in a TM-5000 series power module, owing to both the device being GPIB programmable and having slightly different rear-panel power connections than the xx-5xx modules. An older (TM-500 specific) plugin will work in a TM-5000 power module - just not the other way around.

Maybe. The DC-510 counter appears to be identical to a DC-5010 with the exception of the latter's GPIB interface board.

I got hold of this counter as part of a package deal and only recently decided to test it out. With a 1.00000KHz reference signal applied to the Channel A input, the thing would consistently read 1.13xxxxxxxKHz - and no amount of tweaking the time base frequency adjustment would change this.

The 10MHz ovenized oscillator was verified as outputting the proper waveform, and a suspect pair of ICs (74LS90 and 74LS04) in the downstream divider chain was replaced. I now had a waveform at the 'R' input to the PLL chip (MC4044B) but was it the right one? Seemed to have a lot of ringing associated with it, and the measured amplitude was rather low.

Further, the filter/comparator circuit connected to the PLL - an MC1458 dual op amp - was being driven to the positive supply rail. Its output - normally -0.5 to -8.0v - is applied to a varicap diode which in turn controls the instrument's 320MHz master oscillator. (The MO output is divided by 80 then phase-compared by the MC4044B to the 1MHz time base reference; correction is by alteration of the error voltage. The 4MHz output is then used for a multiplicity of control functions.)

Since the service manual is a little vague on waveform data I had to make a SWAG and figured the MC1458 might be shorted high. Ordered replacement parts, exchanged...nope. Hmmm...maybe the PLL chip 'R' line is shorted low and is drawing the 1MHz reference down? A check with an ohmmeter showed 35 ohms resistance; that'll do it. Desolder the suspect part - an MC4044B - then remove it.

Path from 1MHz reference oscillator feed to 'R' pin foil still shows 35 ohms relative to ground. Damn.

I checked every trace associated with the signal path and could find nothing amiss, at which time a got a pointed soldering tool and scraped between the '1MHz' and 'GND' header pins which tie the timebase and auxiliary/input boards together.

Voila. Infinite resistance.

Plopped the removed PLL IC back into place, reassembled counter...and it worked.

Moral of story: Organic (carbon based) fluxes are hygroscopic...given sufficient time and moisture draw they will become conductive if a thick enough layer is present between adjoining pins or leads.

tekdc5010front.jpg

It's not as compact as a DC-508 and only covers to 350MHz but the '510/'5010 is arguably one of the best modular counters that Tek produced.

Now...let's couple it to a previously mentioned plugin:

tekdp501.jpg

If you wire your TM5xx/TM5xxx power module a certain way - providing rear-panel "Prescale Enable" signals from DP-501 to DC510/DC5010 (or DC509/DC-5009, for that matter) you now have a two or three bay instrument capable of counting to 1.3GHz in both frequency and events domain, and the counter will indicate normal or prescaled operation by virtue of the "GHz" LED. The '510/'5010 units also incorporate math functions which the other Tek counters don't.

I managed to get another DC-5010 at a fire-sale price, and once it's checked out I'm going to look into converting it into a DC-510 - which works in my older power modules.

[N8YX]

Here's another neato plugin, albeit one made by a company other than Tek:

PULSE451APLG.jpg

It's a "programmable MOS/CCD pulse driver".

Sounds convoluted?

In simplistic terms, you apply a TTL (0/+5v) signal to the Input connector and the device converts that to a signal whose logic High and Low levels are set by the bottom two concentric controls. This allows you to control a variety of logic types, depending on how the states are set. Range is +/- 15v. Thus, all major families can be driven properly.

Of course, it makes for a neato RS-232/RS-485 level converter/transmission driver. Just apply your TTL-level ASCII signals to the Input connector and connect the Output to the device under test - after setting the levels.

Transition delay (slew rate) is also adjustable from the front panel. Helpful when playing with state timing and evaluating propagation delay through an array of gates.

These plugins were originally designed to drive charge-coupled device (CCD) imaging arrays hence the flexibility when it comes to setting output parameters - especially rise and fall times.

[K7SGJ]

Here's another neato plugin, albeit one made by a company other than Tek:

PULSE451APLG.jpg

It's a "programmable MOS/CCD pulse driver".

Sounds convoluted?

In simplistic terms, you apply a TTL (0/+5v) signal to the Input connector and the device converts that to a signal whose logic High and Low levels are set by the bottom two concentric controls. This allows you to control a variety of logic types, depending on how the states are set. Range is +/- 15v. Thus, all major families can be driven properly.

Of course, it makes for a neato RS-232/RS-485 level converter/transmission driver. Just apply your TTL-level ASCII signals to the Input connector and connect the Output to the device under test - after setting the levels.

Transition delay (slew rate) is also adjustable from the front panel. Helpful when playing with state timing and evaluating propagation delay through an array of gates.

These plugins were originally designed to drive charge-coupled device (CCD) imaging arrays hence the flexibility when it comes to setting output parameters - especially rise and fall times.

Just out of curiosity, where would you plan to use this? Are you doing some R&D, or do you see an application in your repair endeavors, beyond fixing other Tek modules?

[N8YX]

Just out of curiosity, where would you plan to use this? Are you doing some R&D, or do you see an application in your repair endeavors, beyond fixing other Tek modules?

I'm going to use it (actually, 'them') as a variable level driver in R&D work.

You see, I made a guy an offer he couldn't refuse - - and ended up with several mainframes full of those, the PI-702 Programmable Bias Supply and a PI-451A MOS/CCD driver.

All of them got checked out and calibrated...I'll keep three or four and sell the rest. Likewise with the bias supplies.

I still do a bit of CMOS/TTL engineering work and find myself needing such a beast from time to time. Tek made several pulse generators - PG501/502/503/505/507/508/5110 - and as far as I know, none will allow dynamic level conversion. The PG5110 might, but I'm not paying $1300 for one to find out. Similarly, none allow post-conversion shaping of an outputted ASCII or Baudot data stream.

There are other uses for this thing but I can see it being very handy for working with logic circuitry. Unfortunately, Tek never did make a lot of that type of equipment in the TM5xx/5xxx format. A logic analyzer/word recognizer (which I have), a digital sweep generator (ditto) and a signature analyzer (which I'm looking for) were about it. They also offer a "Setup-Hold Time Calibration Plugin" which I need to get hold of...but not for the solid-gold prices they're currently fetching. Heck, I could design and build one from scratch if I really had to have that function in the lab.