EIP 545A Microwave Counter: power meter upgrade

2014/09/20 Simon 28 Comments

This time, not really a repair, but an upgrade.

Every lab or repair shop dealing with really high frequency circuits needs one: a microwave counter. The EIP 54x series has certainly been (and still is) one of the workhorses of the industry. EIP is now Phase Matrix, still selling counters. But I guess, not an easy tasks – there are quite a few of the EIP counters around, and they do have some little issues with tantalum capacitors, etc, but all in all, the EIP counters are really marvelous instruments, rock solid.

The 545A:

In my own shop, there are two 545A counters (18 GHz, N connector), and also a 548A, which is similar, but covers up to 26.5 GHz (3.5 mm connector). The 548A can go to frequencies as high as 110 GHz, with some external mixers. While these have provided good service over the years, they are all lacking ‘option 2’, the build-in power meter. At microwave frequencies, having a combined power meter and counter has a considerable advantage – no need for changing cables, no need for splitters or couplers – just the push of a button. And not always do you want to connect a circuit under test to a valuable 18 GHz+ spectrum analyzer or measurement receiver…

Fortunately, there are clever people around, at the Green Bay Professional Packet Radio club (“GBPPR”), which kindly provide the ROM images and some circuit modification info on the web – to convert a regular 545A, to an option 2 545A!

Well, first we need to collect some parts, a low capacity Schottky diode (HP 5082-2835, 1 pF max.; original EIP circuit uses a different kind, but really any low capacity Schottky will work), a DM8136N (a comparator/address decoder), a 74LS244 buffer, and an AD7524JN (8 bit DAC) – plus, a 10k resistor. Except for the DM8136N, nothing uncommon at all.

The only tricky part, at least here – the 2532 EPROMs. These aren’t accepted by any of the programmers around, and they need a high programming voltage. So first, put together a little programmer that attaches to an ATmega32L, and with the data transfered by USB.

It’s really crude – I don’t anticipate a lot more 2532 EPROMs that need programming, except for the EIP….
I had some 2532A around which are working perfectly fine, just lower programming voltage. The 2532A/2532 – they can be handled by some 2732 programmers, for reading the contents, with a little socket adapter (2532 and 2732 have different pinouts!) – but strangle, the same programmer doesn’t work for writing to the these old beasts.

The second type, they are 2516 EPROMs, no problem with a 2716 programmer.

After fiddling around with the EPROMs, the little programmer, the adapter sockets – a few hours later – that’s now the set of parts:

These are the boards, A105 and A107, with the parts installed – red rectangles: additional parts; yellow: removed parts (resistors R39 and R40).

I would suggest to put all parts in sockets, like EIP did. Then you can change your counter back, to non-option 2, should you ever want to.

After a bit of soldering – that what we have – a working option 2 EIP 545A.

The power meter can be switched off – if you need the extra digits. Typically, I don’t.

All that is left are the original EIP EPROMs, programmed in San Jose, California. Collectables.

Any questions, please let me know!

Micro-Tel 1295 Precision Attenuation Measurement Receiver (2nd unit)

Micro-Tel 1295 Precision Attenuation Measurement Receiver: 2nd unit!!

2014/09/20 Simon Leave a comment

Recently, I haven’t been acquiring a lot of test equpiment, for my own workshop aka museum, because space here in the US is limited, and carrying all these things over to Germany again in 1 or 2 years will be a hazzle. But this time, I could not resist – a Micro-Tel 1295 receiver, for less than 1 ct. per USD 1980s list price! The parts alone, a 2.33 GHz low noise LO, 2-8 GHz Avantek YIG, 8-18 GHz Avantek YIG, a 2-18 GHz broadband coupler, various microwave mixers and attenuators, all of the best mil-spec quality, well worth it.
Also, it will be a great addition to the precision attenuation test set-up: a dedicated receiver each, for the through and reflected power! And, we can safe one coaxial relais (to switch either through power, or reflected power, to a single receiver), and everything will be faster, by almost a factor of 2. The only downside – another PLL will be required, but well, this is just a matter of a rainy weekend.

It arrived well packaged, no damage, except a missing frequency display bezel (which was easy to source, exact fit), but one thing I did not expect: the paintwork on the upper and lower cover, and side panels, has converted into a mixture of honey and chewing gum, a sticky mess, and dark green! So, first task was to strip off this “paint”, which was pretty easy using some methylated spirits, and engage a bit in spray painting. Hunter’s Green.

See the dark green side panel – covered with sticky paint! Now, it is finally clear to me, why the 1st unit, the 1295 acquired earlier, had been re-painted by his former owner….

Note that the lid and side panels have numerous screws and nuts (more than 100 single pieces!) – quite impressive, how little Micro-Tel had to consider manufacturing cost!

One issue found so far: the IF distribution relais had some intermittent noise – most likely a bad contact somewhere. So took out all the board, cleaned up the edge connectors (all gold plated), and moved the connectors around a few times – and, the issue is gone.

A note on edge connector cleaning: this is best done, from my experience, we some special type of eraser. Don’t use anything harsh, abrasive, or natural rubber. It will either scratch it gold coating, or leave traces of residue behind that isn’t going to improve contact resistance and reliability.
Best suited at vinyl erasers made especially for PET film or tracing paper. Prismacolor Magic Rub.

These crumble a bit, but the vinyl materials absorbs all the dirt, and can be brushed of easily, with an ESD brush.

The unit is overall very clean, just the frequency calibration/display seems to be quite a bit off. This will be the next step, after a thorough warm up.

And, as for the first unit, I will add a parallel to serial converter for the display, same as for my main Micro-Tel 1295, because I don’t want to use the IEEE-488 bus for this device. It will have the same 2400 baud (TTL-level) serial output. Also, a little capacitor will be added, to limit the bandwidth of the Freq Control/YIG driver amp when in external mode – this seems kind of a bug of this device, because the larger bandwidth only increases noise, and the receiver is not build for fast sweeping anyway.

Test Equipment Repairs

HP Agilent Keysight: missing instrument feed mystery

2014/09/20 Simon 1 Comment

There is one thing that has always been inriguing me: why are most of the HP Agilent intruments that I am getting my hands on missing feet?

This is a picture from the recenty fixed HP 8782B. It has 3 feet installed, 1 missing –
HP Part No 5040-7201, 5041-8801, about USD 14 (each!) at Keysight.

Maybe, some sellers take them off, and sell them separately at xbay – same what is happening to the manuals. And in fact, some offers are posted, but by no means enough to explain all the missing feet everywhere.

Also, if equipment is rack mounted, sure, the feet are taken off and stored somewhere in a drawer at one of the big test labs army bases or corporations, but many of the intruments have never been rack-mounted…

Maybe some of the test engineers just can’t let go and hold back at least one of the feet! Well, I’m glad they don’t take out one of the transistors…

So in the end, if you happen to find a bag of these somewhere, please forward them to me…. to put them back where they belong.

HP 8782B Vector Signal Generator

HPAK (HP Agilent now Keysight) 8782B Vector Signal Generator: an easy fix

2014/09/20 Simon Leave a comment

The 8782B is a quite powerful generator to simulate digital signals, not considering the power in dBm, but the support of all the common digital modulation schemes, BPSK, QPSK, 8PSK, 16/64/256 QAM, at exeptional signal integrity.
Frequency range is 0 to 250 MHz, actually, it works to 370 MHz – but this instrument is mainly intended to generate complex IF signals, which can then be used to test the IF chain of a receiver, or, you just add a mixer, another generator for the LO, and a filter, if needed, to get the signal up into the GHz region.

This unit arrived in a HUGE box, it just fits the car!

All frequencies are derived from a HP 10811E ovenized oscillator, which is held by a shock-absorbing mount, and is extremely stable (better than 0.5 ppb per day!), and low phase noise, resulting in below -125 dBc at 1 kHz from the carrier, and below -130 dBc further out.

The symptoms:

(1) The right LCD display is missing partial digits. Seems to be related to the mounting/contact of the LCD with the board.

(2) Output is 30 dB down, but otherwise working just fine.
Quick look at the output amplifier – a real marvel of RF engineering.

(3) It’s in good condition, but will benefit from a bit of de-dusting and cleaning.

The repair:

(1) After taking off the front panel, the LCD is a separate assembly, seems to be made by HP (which test equipment company would nowadays still make their own LCD displays?), and easy to remove. And, the defect, readily found. One of the side bars holding the assembly together and pressing the glass display to the control board, to make contact via an elastic strip, it had slipped off. The root cause – somebody must have used nail clippers to fabricate a piece of acrylic that is used to protect the LCD surface – maybe it was broken or scratched. And with the uneven edges, it caused tension on the display unit, pushing off the side bar.

Well, Let’s just straighten out things and put it back together. No surprise, the LCD is working again just great.

(2) The level control. According to the description of the sender, the output is about 30 dB down. Most likely, something with the attenuator, 08780-60093 = 33322GC, 110 dB, 10 dB step. So, the attenuator, easily removed, and the non-attenuated signal directly fed to the dated but very trustworthy HPAK 8565A spectrum analyzer available here at the bench. Now there is plenty of signal, but it isn’t accuarate at all, even relative steps, like 1 dB steps, give variable response, anything from 0.5 to 2 dB steps… The level spec of the 8782B isn’t all that great, but I know from earlier work with such units, the performance is much better than spec, and 1 dB step, should be 1 dB. Turns out, just a little pitfall, the 8782B has a calibration routine, also for the level, and somehow, this unit seems not to have been calibrated for a while. Well, easy enough, just pushed the “CAL” button, and after a few minutes, you bet, the levels are spot-on, and, within 0.1 dB or so.
Just a quick fix on the attenuator control circuit, and, the level is back.

(3) For cleaning, don’t use anything too harsh on these instruments, they use different plastics and pains compared to earlier HP models. I get good results with about 20-25% isopropylic alcohol (1 part of the alcohol, 3 parts of distilled water).

Some quick tests – all self tests passed OK – and here, a BPSK signal:

Level accuracy, all well within spec (measured with a Micro-Tel 1295):

And, with the 8782B the boring times in the lab are over, Thank’s to the build-in game – everyone will be looking at you, working hard, pushing the buttons!

Ultra-linear ADC

The best solution, most likely: let the nonlinearities (INL) cancel out

2014/09/16 Simon Leave a comment

After putting a bit more thought into this, let’s have a look again at the kind of nonlinearity observed for the ADS1211. We only know three points where there will be no error due to non-linearity: the zero point (because that will be covered by the zero point calibration), and the plus (and minus) voltage, at which the gain calibration is carrier out. The gain will be calibrated both for the positive and negative direction, simply by reversing the same calibration voltage, most likely, about 7.x volts, supplied by a LTZ1000.

Again, from the datasheet:

Now, what if we measure not just with one ADC, but with two, of the same kind, and hopefully, with the same non-linearity, but, with the polarity reversed. I.e., because of the fully differential nature, we can measure, simultaneously, the same voltage, both in the positive and negative direction. Doing this adequately should cancel out most of the (integral) nonlinearity. Furthermore, if use two independent references, for the two ADC we will also gain noise margin – because some of the noise in non-correlated and will cancel out – also, we will acquire the same signal independently, and do the averaging digitally! For non-correlated noise, this means about 3 dB gain, about half a bit!

It only means that finally, we will need to put in 8 ADCs to measure two voltages, but, well, who cares – the given application can handle this, specialized equipment, and no relation to the total cost. And, with some luck, it will result in linearity errors of better than 1 ppm, and 7 digits resolution, with pretty fast data rates.

The solid line: non-linearity of ADC1, the dashed line – non-linearity of ADC2, both ADC are running fully synchronized, same control codes to both ADCs. Two digital outputs – and, one output will be fully inversed, directly in the ADC controller (an ATmega32L), to yield the average of the ADC1 reading, and the complementary of ADC2. Sure the difference to 0 V can also be analyzed, to check how far off the individual values are.
The ATmega32L will also do some decimation, from 60 Hz, to maybe 5 Hz or 1 Hz (independent of the mains frequency), and sent the data to the main controlled, via an optoelectronic isolator (the full ADC section design with fully floating digital and analog grounds). 1200 or 2400 baud will be plenty to get the data out. 60 Hz 6 bytes would be 360 bytes per second, about 3600 baud (need to count start and stop bit), but with decimation, we don’t need any fast couplers, etc.

Sure, this is currently an idea, and will need a closer look, but I assume, it will do the trick, at reasonable cost.
If it works out, maybe we go one step back in the final implementation – the ADS1211 has a 4 to 1 MUX, and rather than sampling simultaneously – we might just give up the noise advantage, and sample consecutively, once with positive polarity, and then, via another channel, in inverse polarity. But hey, Texas Instruments will be happy to sell a few more ADCs.

Finally, not sure if it is better to run both ADC from the same 10 MHz clock, or from separate clocks – some of the jitter induced noise might average out only, if the jitter sources are independent. So many, option, but quite easy to find out!

HP 3562A Dynamic Signal Analyzer

HPAK (HP Agilent now Keysight) 3562A: power supply repair

2014/09/15 Simon 1 Comment

Somehow, all 3562As in the Greater New York City Area seem to fail these days… This one arrived completely dead. Didn’t take a long time to find the issue – a defective A18 power supply assembly.

This power supply design can only be recommended to anyone interested in electronics. It is build with multiple controls, overcurrent, overvoltage protection, and build quality-robustness is not far from being able to power life support devices, or moon landing vehicles from from. Still, this one had failed.

A quick look at the board revealed two blown resistors. Framed red in this snipplet of the schematic.

However, these are the gate drive resistors – there must be a reason why they failed, and this reason was quickly found in the MOSFETs, Q400, Q401, HP part number 1855-0473. Looking around, no real equivalent found in the HP cross reference table. However, an IRF450 (500 V, 12 A, 150 W) appears suitable, and 4 pieces (2 spare, just in case) were easily sourced.

The resistors, R404 and R410, are of more concern. CMF60-64 type. These are 0.5 W resistor fuses (fusible resistors). In contrast to other resistors that can easily start a fire and develop a short when overloaded, these go open circuit, and are flameproof.
Unfortunately, a value of 3k9 seems impossible to source in any reasonable quantity – but 3k3 should work perfectly fine. Digikey offers the Vishay NFR25H series, also 0.5 Watts.

Spare parts

Note – why did HP use a fusible resistor? Why not replace with a regular resistor? Please, never even consider it, unless you it’s a client that doesn’t pay the bill… it’s a serious risk of fire, and it is not just a bodge but an unacceptable safety hazard.

That’s why… the red trace is 400 VDC at multiple Amps, and imagine the Q40x have a gate-drain short…

Didn’t take long to replace the IRF450s and the two resistors – and, to everyone’s full satisfaction, the 3562A powers up, no issue. Not sure what caused the defect in the first place – maybe some overvoltage in the power line? We might never find out.

Ultra-linear ADC

ADS1211 evaluation: a bit of shielding goes a long way…

2014/09/14 Simon Leave a comment

Following-up on earlier analysis, the ADC performance observed at small decimation (like 1 s averages, with 60 Hz data rate – 60x decimation) left something to be desired – quite a few bits lost due to mains (and/or other) sinusoidal noise.

Look at 1 hour worth of data (60x decimated; each sample: 1 s worth of data), at higher magnification than before – ticks every 15 minutes:

Obiously, there are some bursts. And these are almost certainly not related to the ADC or anything internal to the circuit. Maybe the power supply, but added quite a few decoupling capacitors…

So, if the noise source is external, a bit of shielding might help – great to have a little metal box (even a steel box) at hand!

Note the 2 BNC cables – rather than one – feeding the test signals from a fully-differential source, to avoid ground loops.

Well, not much to add! Amazing what a little metal box can do!

Ultra-linear ADC

Studying the local characteristics of the ADS1211: ADC differential nonlinearity, missing codes

2014/09/14 Simon Leave a comment

To qualify the ADS1211 for the given application, or at least, to gain some confidence in it, a test – not the for overall non-linearity (i.e., non-linearity over the full range, aka integral nonlinearity INL), but for the more detailled view at the ADC’s precision.

Local deviation of an ADC from linearity are called differential linearity, and this can be some slight deviation, or can go so far that there are even “missing codes”. A missing code is caused by a local non-linearity that is larger than 1 LSB, to the ADC will jump 2 steps, even if the voltage is only increased by 1 LSB equivalent.

First, the test setup: still the ADS1211, running at 4 MHz, 16 turbo mode, 60 Hz data rate. Connected by fully-differential coax to a (floating) source, an HPAK 8904A signal generator. This is programmed for a 5 DV output, with 20 mVpp (intentional) sine ripple, 13 Hz. The selection of the frequency is rather critical, don’t let it be anywhere close to a subharmonic or harmonic of the data rate!
The HPAK 8904A is actually really great for this purpose, you can add and mix any signals, up to 4 channels, and modulations, as desired, into one channel!

Alternatively, you could feed DC-biased noise, but these noise signals can be troublesome, and you never now what to expected in terms of amplitude, flatness, etc, unless you have really specialized gear.

Having everything set up, several hours of data were collected. Virtually no drift, so the DC component-the average ADC code (nearest integer) was subtracted from the data, and the results analyzed.

Full data, +-2000 counts is more or less +-10 mV (20 mVpp), as expected. 1 LSB is about 4 µV. There is dot for every count, even if no sample was recorded, at the given count (then, the dot is at 0 samples…).

The probability density function (PDF) corresponds to that of a sine function. That’s a good start.
Some key observations – there seem to be 3 “populations” of sample counts – codes that are “0”, i.e., missing codes; codes that have counts that are somewhat in-between (the majority), and double-counting codes. This needs some more investigation.

Taking all these data, and the know PDF of sine (of the form, 1/(x*(1-x), “bathtub curve”), the PDF was fit to the data, using least squares.

Green line shows the fit-this makes sense, and the residuals were calculated.

We are only interested in the center part, where the errors due to drift are minimal. A close up:

We can cleary see a pattern: DxMMxxMMxxMMxxMMDxMMxxMMxxMMxxMMD…
D – double code, M – missing, x – intermediate.

What seems dramatic, it’s acutally not. There aren’t any deviations more than +-1 LSB, and there will be noise and averaging anyway, to get beyond even 22 bit resolution.

Harrison HP 6202B Power Supply

Harrison HP 6202B Power Supply – revisited

2014/09/14 Simon Leave a comment

After doing a quick repair yesterday – a sleepless night, because of the out of spec ripple. Well, not quite that bad. But typically, all HP instruments hold up to standard, it would be a big surprise if this one can’t be brought back into spec.

First, desoldered all of the capacitors (they used a lot of solder!), and checked them – surprisingly enough, they still work perfectly fine, all 1967 vintage.
Next item, the recifier diodes – some might be leaky. Therefore, just replaced them with “modern” 1N4004. But this didn’t cure the ripple.

Next item, the 6202B has a reference circuit, and, the output had 10 mV noise – ripple. That’s far too much, and did not fully feed through to the output.
What caused this issue? – the defective capacitor (see earlier post) is the filter cap for the reference. Seems the replacement cap doesn’t give the same filtering – just added a bigger capacitor in parallel.

The result – at about 50% load (2 mV per DIV, sorry, don’t have a better scope here)

About 0.8 mVpp, estimated 0.25-0.3 mVrms. Amazing. Fully working again.

Ultra-linear ADC

Noise: external and internal

2014/09/13 Simon Leave a comment

A quick – 9 hour test – of the ADS1211: at 60 Hz data rate, 16x turbo mode. About 2 million samples.

According to the datasheet, every sample will yield about 22 bits noise free data, according to the datasheet (red circle shows setting used):

That’s what has been obtained.

Clearly visible, higher density of codes at the left and right end. Really suspicious – almost certainly, nothing else than a bit of mains feed-through, about 175 µVrms. Seems we get >20 bits, more or less, otherwise we would not be able to see the distribution (note that some counts have a positive deviation – due to non-linearity!).
But all this, with some 60 Hz noise on top…
Assuming that this noise is constant, it can be eliminated either by futher digital filtering, or by averaging/further decimation of the data, which will be done anyway. As a rule of thumb, decimation by half will give an extra 0.5 bit of data, for random noise – and a bit more for constant 60 Hz.

Here, a quick look at the decimated data (note that the sample axis ticks correspond to 2 hours time intervals):

In these diagrams, “full scale” corresponds to 20 V – the current configuration can hande up to 60 V differential signal, at 1 LSB equal 4 µV. So there is still room for improvement of resolution, like 1 LSB equal 2 µV – but this only provides useful data, if we can get noise down well below 2 µV, which will be tough anyway.

Also, quick statistical analysis, of the 10 s average – 600 sample average data:

Standard deviation, 0.71 LSB equals 0.71 LSB RMS noise, equals 3 µV. Quite reasonable!
Still visible are the two maxima of the distribution, due to the mains noise.

Comparing to a random-noise based decimation-improved resolution (noise free bits), it seems that the converter is yielding about 20 noise free bits, at 60 Hz data rate. Not quite the 22 bits mentioned in the datasheet. This is not surprising, the last 2 bits, at the fast rate, we will only be able to get this with a better, ultra low noise reference, a low-noise bias supply, and low noise analog power supply (currently using the build-in reference, and build-in bias supply, and 5 V supply from USB bus…). But fair enough, about 24 noise free bits (7.5 digits!), at 10 s averaging, and 9 hours zero point drift of less than 0.4 ppm, this might already be good enough.

SimonsDialogs

EIP 545A Microwave Counter: power meter upgrade

Micro-Tel 1295 Precision Attenuation Measurement Receiver: 2nd unit!!

HP Agilent Keysight: missing instrument feed mystery

HPAK (HP Agilent now Keysight) 8782B Vector Signal Generator: an easy fix

The best solution, most likely: let the nonlinearities (INL) cancel out

HPAK (HP Agilent now Keysight) 3562A: power supply repair

ADS1211 evaluation: a bit of shielding goes a long way…

Studying the local characteristics of the ADS1211: ADC differential nonlinearity, missing codes

Harrison HP 6202B Power Supply – revisited

Noise: external and internal

SimonsDialogs – A wild collection of random thoughts, observations and learnings. Presented by Simon.