Category Archives: Various

Dual CV 1600 Stereo Amplifier: a real HiFi classic with some capacitor smoke, and a very large dropout voltage regulator

This one is a real gem, a stereo amplifier, Made in Germany, and nicely constructed in the 1970s, using mostly discrete parts.

cv1600 front

Some issues are common to all old amplifier, like, defective switches, aged contacts, and so on, but these can be fixed with good contact spray, or by (mechanical) repairs. For the CV 1600, most common fault is the X2 mains capacitor, 47 nF, which will eventually turn into smoke and stench. Be sure to replace this part, if you have a CV 1600.

The 47 nF X2 rated cap is located on a fuse board, and to get access, you need to remove all the transistor wires. This is best done with great care, and without damaging the wires, otherwise, it is quite laborious to connect everything back up again.


This CV 1600 also had another issue, no signal from the distribution amplifier board. Some first check showed issues with the +15 V rail being stuck at ~7V. Probably a defective Tantalum or other capacitor? Probably not – not much current flowing either, so it is not related to a short at the output.


Inspecting the regulator – it is a TO-202 LM341p15, pretty rare nowadays.

dav dav[/caption]

No heatsink on the regulator, so I got a bit worried with the dissipation of TO-202 vs. TO-220 devices – all I have in stock are TO-220 regulators, LM78xx series.

cv1600 to202 vs to220

lm341p15 to202

lm7815 to220 heat resistance

After inspection of the datasheets – nothing to worry about. Pretty similar heat resistance, junction to ambient. Voltage drop is about 10 V, current roughly 50 mA, so, 0.5 W – roughly 30 K temperature rise.

A quick test with a 5.6 Ohms, 250 W dummy load showed no further issues (except an incorrectly installed signal cable, probably from an earlier attempt of someone else to fix this unit?).

Finally, some performance data of the CV 1600.

cv1600 data

HP 6115A Precision Power Supply: repair complete!

No repair can proceed, without sufficient time, and without the right spare parts. Time was very much occupied by other business recently, spares took time to ship from Greece to Germany… the 2N6211 transistors.

Here they are – the 2N3442 are China-made (ISC) transistors of rather recent production, the 2N6211 date back to 1992.
Not much to write about the further repair, mounted the transistors on the heat sink, soldered-on the cables, replaced the Zener diodes of the main board (series regulator bias), and switched the 6115A on. Success! Some minor calibration of the panel meter. Other than that, all in good shape.


Various things can be measured to verify the correct operation of the 6115A, here just a quick test checking all the ranges, linearity, and deviations. In short, it is very much more accurate than the 0.025% + 1 mV output accuracy specification.




After a full cool-down, checked the stability/drift after a cold start, with the output programmed to 10 volts. It is ramping up nicely, with some very minor “instability” during the first hour, but then stabilizing to almost perfect level. This is with no load. Sure it may degrade a bit under full load, fair enough (horizontal axis shows measurement number, period is about 0.3 s per measurement).


Finally, this is the working precision supply.


HP 6115A Precision Power Supply: not just one defect mixed with a lot of precision

The HP 6115A is a really great power supply, 0-50 V @0.8 Amp, 50-100 V @0.4 Amp, 0.0005% load and line regulation, 0.01% current regulation, 100 uV p-p ripple, 0.0015% drift over 8 hours, 0.025 + 1 mV accuaracy of output voltage, all in all, challenging design objectives still today. Unfortunately, the unit discussed here has not any of these characteristics, it is dead, and missing the current adjustment pot.


At least, it is a reasonable clean unit, and at a first glance, nothing major, like a completely melted board or smoking transformer. Judging from the soldering, someone already tried to fix it but gave up mid-way. Sure, I will not give up with this supply too soon.

First things first – the current pot, a 10 turn 1 k, HP part 2100-1864 (Bourns 3540S): missing. Looking around in my parts collection – but no luck. At the very bottom of a stack of old electronics junk, a WTW 610 pH meter from the 1970s. This is used to convert pH electrode signals, to proper voltages, but even more important, it has 2 pcs 1k Helipot 7276 series 1 k pots. These are 20 ppm tempco, even better than the Bourns (50 ppm).




With the 1 kOhm pot fitted, still no function. About 60 V at the output, and the current limit LED lit, regardless of current or voltage setting.

Looking around inside – a few issues found. There are several uncommon parts, like, a STB523 = 1N4830 voltage stabilizer, which is more or less a stack of 3 Si diodes. These parts are not commonly available anymore, so I replaced the defect regulator diode with a series assembly of three 1N4148 diodes.



Found a few more issues, several blown Zener diodes, all around the Q1-Q4 transistors. This is not good, because it may indicate some blown power transistors. And if fact, it did not take long to find out that Q1 has a full B-E-C short, and Q4 is C-B short, E-C, E-B open. No wonder that this disturbed the bias network Zeners, VR3, VR4.


The transistors, 2N3442 NPN (Q1-Q3), and 2N6211 (PNP, high voltage power TO-66), at least the latter, not quite common – on order from an xbay seller in Greece, 5 EUR a piece of old and obsolete part, OK!
The power transistor board was an awful mix of bad soldering and flux, finally, cleaned and most of the solder removed.


Found another defect – no voltage on C8, one of the main capacitors. Reason: a blown trace, from tap 16 of the transformer. Temporary fix with a yellow wire….


After some repairs, at least the basic voltages and supplies are up and working again, all capacitors tested, and working fine. Also reviewed the regulator board and its voltages (not shown in below diagram, red cross means dead part removed, green cross means part absent but tested good), all is fine and working.


Note the working principle of the series pass regulator – it is a dual range setup, with a high voltage regulator, Q1 and the low voltage regulator Q2/Q3, all driven by the common Q4. Diodes CR11 and CR12 (a dual-diode element) is directing the current from the appropriate transformer DC supply (2 equal DC voltages of about 80 Volts are generated from two separate windings).


Apart from a 1N829 0.0005% tempco temperature compensate diode, there is another remarkable part used in the circuit, a 10 ppm 10 turn trimmer – not quite cheap, and still available today!


This is the current state of the instrument, waiting for the spare transistors, before I can put it back together, and hopefully, put it back in service.


HP 83572A RF Plug-in 26.5-40 GHz: only a fuse away from the highest frequencies

Not one of the most preferred things to repair – a rather rare 26.5-40 GHz sweeper plug-in, not producing any output. Despite its rather simple function as a signal source, typically, not easy to fix if any of the microwave parts are faulty. New, close to 18 kUSD, so it is nothing you can easily replace from a hobby budget, and 26.5-40 GHz sources are not easy to come by.



Inside, all full of heatsinks, and a few waveguides.


The modulator.


The YTO. Not many companies around that can manufacture such devices.


First thing to test, with no output present – the YIG oscillator. This has two main items: the bias supply, which is more or less just a variable voltage power supply which is tuned along with the frequency sweep. Secondly, the main tuning coil current, providing the magnetic field for the YTO. Checked both – and found the bias supply at 0 Volts. No wonder there is no output.


Upon inspection of the schematic, I noticed the fuse, which is rather hidden down on the motherboard. And, it was blown. No idea why – maybe just because of its age? Sure enough, HP did not use just any ordinary fuse, but a BUSS GMW model.


USD 9 per piece – that’s a steep price for a fuse.


Cut the fuse open, and connected a 5×20 mm European style fuse. All protected by a piece of shrink tubing.


Well, an about 1 hour later, the YTO is oscillating again, and you can see a nice and strong signal, well over 90 dB useful range, to test attenuators, or whatever 26.5-40 GHz device you want to put to test.



EIP 545A Microwave Frequency Counter: another dead tantalum repair

It’s not the first fix of an EIP counter here, and I have to say, these units are still very useful tools around any RF and microwave workshop, despite their age. The unit under repair, judging from the date code of the parts, is about 25 years old.

The symptoms – the EIP is just not counting. Showing 000000 zeros in the display, but no count. To confirm, you can use the build-in test function 01, by pushing test-0-1, and normally the counter will show “200 000 000”. But no indication of any counting activity for the current unit.

First thing you do, checked all the rails, and turns out the 12 V rail is dead. Starting from the right, removed all boards inside. And soon the defect was traced to the A107 board, gate generator assy.

Checking with the schematic, there are several ceramic and one tantalum cap.


Desoldered the tantalum, and it is no cap any more, just a short.
With all the various tantalum shorts repaired in the past years, there seem to be to cases, some tantalum go fully short and activate the power supply protection circuit, with little powder dissipated over the tantalum. This is the good case. Some other tantalum seem to develop a short with some resistance, leading to considerable power dissipated in the cap, causing stench and risk of fire. The latter more seems to be more common with SMD tantalums, for reasons unknown. Also, the tantalum story is one of the many reasons why you should design current limit circuits, and power supply protection circuits, even in low power equipment, especially, if there are any valuable components that you might want to protect from a power supply failure.



The 10 uF replaced, with a new orange cap, but same “drop” style.


After carefully installing the A107 assy back into the counter (take care not the damage any of the wires), all is good. Counting at 200 MHz with the test function 01 activated.

A Thermal Fuse, and HP 10811-60111 Repair

Usually, I don’t care much about high precision oscillator options being fitted to frequency counters, etc., because in the lab, any critical equipment is anyway connected to an external well-controlled 10 MHz reference, locked to DCF77. However, this time I need to install a OCXO (HP 10811) in a HP 5335A counter for service outside of the lab.

The only thing that needs to be done is to remove a jumper on the board of the 5335A (see red box in picture below), and mount the 10811 in the slot already prepared for the OCXO inside.

ocxo 5335a jumper

While such installation is fairly straightforward, it turned out to take more time than expected – simply because of the OCXO not showing any stable output signal.

ocxo 10811-60111

After a few quick tests, the cuprit was found, a defective (open) thermal fuse. This is apparently a quite common issue for the 10811 oscillators, and you might get away with just putting in a wire jumper. However, I didn’t want to take any risk of overheating in case of a failure of the 25+ years old OCXO circuits. An exact match for the thermal fuse could not be found, so just soldered in (very carefully, cooling the case and leads!) a 10 Amp 109 degC fuse.

ocxo fuse picture

This is the OCXO with the new fuse installed.

ocxo new thermal fuse

This style of fuse as a non-insulated outer shell, so a shrink tubing sleve serves as insulation.

ocxo insul sleve

Finally, a note found in a datasheet of a common thermal fuse – it clearly states that lifetime will be limited when operating the fuse to close to the cut-off temperature. So clearly, thermal fuses are not the best protective mechanism for the OCXO case. Maybe better would be a bimetallic switch (self-resetting, but at least no subject to any significant aging), or some other device like a PTC.

Sure, we can slightly blame the HP engineers, because it is stated on most thermal fuse datasheets, like the one below, that the operation temp limit should be about 30 degC less than the cut-off, which is not quite the case for the 10811 OCXO. 80 to 84 deg C operation, 109 degC fuse cut-off.

ocxo storage temp

ocxo thermal fuse

HP 3325B A26 Assembly: low byte SRAM, and an incorrect signature

A quick comment on the 3325B service manual – Thanks to a kind contributor.

3325b kernel sa test u27 pin10

The kernel SA test is a bit flaky on A26U27 PIN 10. This is an inverter output, and can’t be same as the input… correct signature is A70F. Appears to be a misprint in the HP service manual!

3325b a26u27 74as04

In the present case – the culprit was one of the SRAMs, the lower byte. Fortunately, not really a rare part, and easily sourced and replaced, once you know that this is the circuit at fault. You might want to check this very carefully, because there are multiple versions of the 3325A/3325B control boards.

3325b srams

HP 436A Power Meter: smoke and stench – X-rated cap failure

By coincidence, another HP 436A power meter – this one, emitting smoke and terrible stench! The culprit was easily found, a defective X-rated cap. One of the known-bad epoxy covered capacitors that tend to blow after about 30 or 40 years of service.

436a x2 capacitor 100 n

The residue, oily stuff, terrible smell. Use plenty isopropanol or methylated spirits to clean – otherwise, the stench will stay with the instrument for years, and I can’t say that it is a healthy smell.

436a 100n oily

The cap is of the well-known PME271M series. Still available, but hopefully, with improved construction.

436a pme 271 m 610

436a pme271 series

A replacement is easily found – taken from an old switchmode power supply. Make sure to take a “X2” cap, not an ordinary cap. Only X2 caps are specified for mains voltage service, and self-exinguishing, anything else will present a major fire hazard, don’t compromise on the choice of capacitor!

436a 100n x2 replacement

Fix complete – new cap soldered in, and insulated with some electrical tape. In general, I tend to avoid electrical tape where possible, but in this case, it appears to be the only viable solution.

436a fix complete

HP 436A Power Meter: a strange analog ground issue

This power meter had been received with strange defect, a permanent overrange error, irrespective of any settings or input to the sensor. Sure enough, in most cases, this would be because of a dead sensor – but not here.
The 436A is a really simple instrument, at first glance, but with its design dating back over 40 years (mid-1970s), it has a remarkable complex design to achieve the A/D conversion, and to use something close to a CPU, at the time, called a state controller.

What was wrong with this unit? Something with the analog ground driver.

Checking the A2 and A3 assemblies, it turned out that the analog ground was floating, at about -6 V. Strange! And, simple enough, grounding the analog ground on either A2 or A3 solved the issue! For a temporary fix, a wire was added, from the board edge connector, to chassis ground. Need to look at the analog ground driver…

436a analog ground wire

Using chassis ground for general grounding – an indication of the dated design, and some of these board use 3 or 4 separate grounding path to keep noise down…

436a analog gnd schematic

After this fix, working again (still need to check out what it wrong with the analog ground driver).

Update: found the issue – lower right and corner of above diagram, this is the analog ground driver (also supplying analog ground to the A3 assembly (via mother board) – transistor Q1 found dead, a 1854-0003 (which is equivalent to 1854-0637, JEDEC 2n2219A, or any other ordinary 0.8 W NPN transistor).

436a 1854-0003 2n2219a

Soldered in a 2n2219A, and removed the temporary ground wire. Fix done.

436a analog ground circuit

After a full calibration and extended test, the instrument is rock stable, both for zero point, and 1 mW input. Also checked linearity, and it appears to be better than any means available here to check… most likely, better than 0.1 dB.

Output of the 50 MHz 1 mW cal source – cross checked with a calibrated HP power meter, 437B , and in agreement within 0.01 dB – good enough!

436a pwr meter working

Some other issue with this unit – a stuck analog indicator. After disassembling the front panel, used a razor blade to open up the plastic case of the indicator, and some mechanical adjustment of the inner workings fixed the issue.

436a analog indicator

436a front panel

The 7 segment decoders, these use heat transfer compound, for some pretty unusual way for HP design – being pushed vs. the front panel for cooling. To make sure these stay cool, I added some fresh white stuff.

436a dm9374 7seg decoder driver latch

The decoders are quite remarkable anyway, for their time – these are latching decoders with constant current output, high level integration for the early 70s….

Kebaili CPG-500 Current Pulse Plating: reverse pulse plating power supply

A rather rare guest in the workshop, a specialized power supply, for reverse pulse plating. Most of these are huge boxes used in the plating industry, with current rated in 10s of Amps, but this device is made for the semiconductor field, for depositing metal on silicon, for MEMS (micro-electro-mechanical) devices, etc. It’s a rather lightweight device, in a handheld case, powered by an external DC power supply.

cpg-500 kebaili

The internals show a construction which is more like a prototype than a fully engineered device. Epoxy glue is used to hold in the buttons, and Capton tape at various places. Well, some prototypes last forever, and it is certainly build to a good standard. The folks building it were very concerned about their design it seems, and removed all the marking from the parts used! Keep in mind, this is just a pulse current supply – why all the effort, and why make it so difficult to repair! Even more striking is the fact that the company manufacturing these is now out of business, with no information available whatsover, and no service provided (at least, I was able to obtain the user manual from the owner of the device). Please, leave the markings on the circuits, they are there for good reason – to let others fix the devices, when your greed and protective nature has ruined your business.

cpg-500 board

A few words about reverse pulse plating. You all know the through-hole plated circuits boards, plated vias, etc. These are made by electrochemically depositing copper, a method which Faraday and others invented long ago, but only in recent decades refinements were made to the current profiles to get nicely through-plated (and also filled) holes. Being able to deposit copper in cavities of any shape, etched into other materials, opens up a tremendous number of applications (e.g., complicated patters can be etched into silicon using the well-established methods of the semiconductor industry, then filled with copped – after removal of the silicon by selective etching, the small copper parts remain, resembling all the detail etched into the silicon).

Here, the general pulse sequence: forward current pulse, reverse current pulse, idle time (no current).

cpg-500 reverse current plating generic

A plating example (not from my lab), (a) shows a not properly plated through-hole (cavity in center) resulting from conventional plating; (b) and (c) are pulse plated – one can see how nicely the copper is growing, even in the center. Needless to say, all can be adjusted by selecting the right mix of chemicals, and by optimizing the current levels and pulse duration for the desired results (filled hole, filled cavity, wall-plated hole, and so on). Typically pulse times are milliseconds. But can be 10s or 100s of milliseconds, in cases.

cpg-500 plating example generic

The CPG-500 has 3 current ranges, quite a flexible device, to cover applications from micrometers to square centimeters…. resolution is about 1:4000, about 12 effective bit.

cpg-500 current ranges

After some repair, no spectacular enought to write about, these are the output signals. Pretty clear how it works – lower trace is forward current, upper trace reverse current – during operation, the current sources is switched from one output to the other and back, grounding one of the outputs at a time. When switched off/when plating is finished, both outputs are switched off/high impedance, with no current flowing. Orange lines are 0 V ground potential of the power supply.

cpg-500 current test

Because of the way this bipolar operation is realized (by unipolar switching of current), probing the current requires a few tricks. A 100 Ohms resistor inserted in the current path (in series with an amp meter), and two oscilloscope probes.

cpg-500 current probing

The scope is then used in ‘add inverse’ mode to subtract the two signals, resulting in the display of the actual forward and reverse currents, as one trace.

cpg-500 plating current example

This is a quite typical waveform, the reverse current is 1.5-3x the forward current, at about 2-10% of the duty cycle. This example as 1 mA forward current, 3 mA reverse current, 40 ms forward, 2 ms reverse pulse (i.e., current ratio is 3, and time ratio is 20). This results in an average current of (40*1-2*3)/42=0.8095 mA. An, not really surprisingly, this is what the amp meter shows.

cpg-500 average current

Call it fixed!