M2 RF detectors, level meters, attenuators,
RF termination (dummy loads),signal dividers

m1. LC- and Q-meters
m3. RF signal and power generators
m11. Griddipmeters and xtal testers
m12. RF power- and VSWR-meters
m21 Norwegian instruments
m22. Old measuring instruments
m23. Signal-to-noise-meter

#2.1) Surplus attenuators

Muirhead & Co No. 355748, it has 0-61.5dB in 0.5dB steps (75 ohm).

Problem with attenuators is that when you need them, you can't find sufficient amount of them. I've often needed 3 for some experiments, and 75ohm attenuators in a 50 ohm system does not cause much inaccuracies, so when you have checked it and are aware of the possible problems you just don't care so much.

Another argument is that I use a lot of available cables and equipment from work which are made for 75ohm systems, and it saves me a lot of trouble to use what is around instead of buying what you really don't need.

Inside look of the Muirhead 75W attenuator (boatanchor). It is well screened between the sections, but isn't much looking like an instrument for VHF, on the other hand it seems to be good - at least for the lower part of VHF range. It came with some Marelli radio link equipment and I had to change the connectors to BNC. Have passed some to other amateur radio experimenter and they are quite satisfied with using them for 50W. But again you need to be an electronics engineer - with 30 years experience to understand how easy it is!

#2.2) Telonic Model TG-975 (First owner: Tandbergs Radiofabrikk, year: 1965)

Another 75W (0-102dB) attenuator I bought in 1980 because it looks so easy to repair. Has used it on VHF and 70cm (50W), when you are aware of it it is only marginal difference between a 50ohm and 75ohm instrument. The switches are mounted between the different rooms with the resistor through holes.

#2.3) Danbridge Decade Attenuator type DA3HS/D; 600W, 0-111dB in 0.1dB steps.
Decade resistors and decade attenuators are among those items I regard as waste or time and money, but I got it from work, and have never used it, or missed it. Such things are usually believed to have a value by persons who don't understand much. I wouldn't use a decade restance unit to find a resistor value! This unit is better for next junk-sale.

(S.: atenaters)


#2.4) Practical VHF/UHF Dummyload to build

Building termination with standard components is easier than people think!
But, usually experience is better than thinking.

Practical dummy-loads mounted on N-connectors. Using male connector it can be connected
directly on the Bird Model 43 wattmeter. it uses 6x 300W 1W Draloric LCA0414 carbon resistors. To ease soldering a timmed iron plate is attached to the connector first.

The Rohde&Schwartz team offered to measure it at a their stand at the Weinheim UKW Tagung, the one with male connector has almost 20dB return loss at 1300MHz, - not bad for standard type carbon resistors! Only had one available.

#2.5) Power Attenuator (25W 30dB)

Rohde&Schwarts RBD 600MHz 30dB attenuator uses discrete power resistors, note the frequency limit, it is professional equipment! The attenuator is original 60 ohm, so I have modified it with some extra resistors to 50 ohm. The original Dezi-fix connectors were easily changed to BNC.

#2.6) more power attenuators

Some power attenuators used for experiments on HF

High-level PIN-diode attenuators, see page c11

#2.7 RF Level meter using Philips TDA1576

10a) TDA1576 (Philips) as level meter, has 70dB range. But using the internal meter offset adjustment circuit upset the range, so it seems better to use an outside circuit. Also useful as quasi-synchronous quadrature AM-demodulator, but it is overloaded with larger than 100mV input.

Bird's nest using TDA1576 as a general purpose detector for 10kHz-15MHz

#2.8 LA7MI linear scale RF-mV meter for 0.01-1500MHz.
Believe it measured better than 10% accuracy to a linear scale on all ranges except the lowest (30mV) up to 500MHz. It was not intended to by LA7MI to have this range since strong transmitters in the neighbourhood gave over 30mV deflection with the high impedance probe, but I added it in my version since I mainly measure 50 ohm circuits via coax cable, and only very seldom use a high impedance probe. Also used it with small coax loop to align local osc for 2320MHz transverter.

RF probe for RF mV-meter. Ground on two-sided PCB laminate component side is connected to the screen side below.
The board patterns is fabricated using hacksaw. The construction is so self evident that I believe you don't need to have the words translated, it was only a point in the first amateur radio magazine it was presented, and as such not shown in DUBUS.
jordplan=groundplane, probespiss loddes her =probe tip soldered here, jordkrokodille = crocodile clip to ground.

See also the other alternative instrument described below.


#2.9) RF detectors using 1N4148 type silicon diodes.
An extra diode is used to forward bias the detector. It has relatively low impedance and decoupling is not usually necessary. Using a transistor instead of the diode it is easier to optimize the forward bias voltage, but the voltage drop over the diode is current dependent, and still with 250mV forward drop it draws some current. Another problem is temperature stability.

Supply voltage Base voltage Collector voltage DC output voltage
9v 643 518mV 44
12v 659 476 34
15v 649mV 438mV 26

It is important that the supply voltage is stable to use the circuit shown above

RF input (450kHz) RMS Detector output (2nd circuit)
-- +30mV
25mV +33mV
80mV +58mV
250mV +300mV
400mV +625mV
800mV 1.568V

The detector is not linear, but may be useful for several non-accurate purposes

c42 Variable bandwidth xtal filter with constant gain using negative impedance amplifier
- (a technique used in HP141 and similar instruments)

#2.10) Very old LF selective level meter (Messkoffer)

TEKADE Träger-Frequenz Meßkoffer 5-155kHz (later German term is Selektiver Pegelmesser), Suppose this is the first selective voltmeter ever made. They were used to measure levels on multi-channel carrier frequency systems. One such instrument were placed in Kristiansand, another in Arendal since it was subsea coaxial cables to Denmark, believe it was 8 or 15-channel systems (ME8 & MG15) during the war and many years after. See Die deutschen Funknachrichtenanlagen Band2 "Der Zweite Weltkrieg" pg139

See chapter M22 for more info about old instruments.

#2.11) Siemens D2006 Selective level meter (Selektiver Pegelmesser)

The advantage with this instrument is that it is still easy to repair and modify.

Sidetone modification.
The problem with many of those Selective Level meters is that the pitch is meant for somebody else, but not hams.
It is far too high for CW reception, for this unit it is 1.5kHz (some other instruments may have 1850Hz). So you cannot use the lowest bandwidth to listen to a CW station, only for measurements, and must tune to lower pitch using wider bandwidth. When I listened to SAQ on 17.2kHz (Alexandersson Generator) last summer, the signal to noise was only 6dB or less, but I later discovered that it was some strong noise sources
just above this frequency, and if I could use the 80Hz bandwidth it would improve 30dB. It was therefore a good idea to open the handbook and start interpreting the German text. Soon it was discovered the solution here is quite simple, all the modifications are made behind the front plate, none on the actual board, you remove some components and install a swith with at least 4 posititions and a small board for trimmer capacitors with some fixed values in parallel. It was difficult - even when reading the book, to understand the proper way to open the instrument. Was it neccessary to take off the front? Yes, it was, and it seemed rather difficult first, but wasn't so difficult, when you learned about it, and I needed to remove it later, because I forgot some screws.....

The board S45035-D5142-..(Ausgangsteil) has a simple product detector and BFO. The BFO frequency is tuned with 2250pF and it is shifted by S4, possibly shifting it 1.85 below and above 100kHz as the switch is set to left or right. when the 250pF is grounded it is LOW and when it has the external capacitors in series it is HIGH. The unmarked capacitors above the switch behind the front are 25pF styroflex and 25pf air trimmer, I measured 41pF, but don't always rely on the instrument, and to this figure you must add some 50pF for the PTFE-cable.
Remove all plates and screws and when you still can't find a way to get into the box, you must start reading the instruction book, Chapter 4, MAINTENANCE - only to discover that you can just as well put back some of the screws again and remove the upper unbrako type screws instead - to slide the upper unit backwards. It was a fight to remove the front plate, but it went after all. Careful with all those terribly small 2.5mm screws, nuts and washers!

Considered using varicap diode (BB112 or BB130), but it must be shorted for USB, it demanded a potmeter with an Normally-Open-switch and they are rare, so I decided use a switch with 4 positions instead.

Connected 3 trimmers and fixed value capacitors on a bracket with pcb laminate where the old capacitors had been.
The 4 positions are now:
1. Shorted coax cable as earlier CCW with 1850Hz pitch for 80Hz filter.
2. 470pf+270pf + 100pf trimmer - to tune one sideband with 500-800Hz pitch
3. 56pf + 100pF trimmer, to tune the opposite sideband with 500-800Hz pitch
4. 22pF + 20pf trimmer (or 10pf fixed and 40pf trimmer) for CW as before

used a six position switch, but couldn't think of the need to use more than 4 positions, didn't find any small swiches with long shaft, so the nut must protrude through the front plate, so the hole must be widened a little for the nut.

The audio output level is high enought to drive a 300ohm dynamic phone inset (+1dBm or 700mV)

Servicing Siemens D2006 Level Meter, when you have unscrewed the proper screws amongst the many which seem much more likely to choose...


Siemens D2006 Level Meter.
Block circuit

#2.13) Active hi-Z RF probe for use with D2006 or other instrument

The probe under check connected to Siemens D2006 selective level meter (or other similar instrument)

Circuit diagram for the probe. 1dB variation 50kHz-18MHz using Siemens D2006 Level meter (75 ohm), probe voltage loss: ca 1.5dB. The input coupling-capacitor is 100pF.

An alternative is the Maxim MAX4005 with frequency limit of 950MHz, but I needed the circuit today and don't want to wait for something I am not really sure if could find or how long it takes to get it. This device requires +/-5V supply.

>.905. see block diagram for D2006 Pegelmesser (too large to display on this page)


General purpose high input impedance amplifier was designed by LA7MI in 1973 for use as active antenna. It is important not to use more gain than neccessary, also the positive feedback to the input, should be done with care, and avoided if not definitively neccessary. We used it for a while, but I had some intermodulation problems with a local BC-station, it was later used with a general coverage frame antenna covering 0.5-4MHz without any problems. It is a simple circuit for some test purposes


Linear RF detector
I've wondered what is the important rules to make a linear detector with conventional technique and germanium diodes It is described in Wandel&Goltermann SPM-3. The upper transistor operates as constant current source and at the same time balances the collector voltage for the first stage.

RF detector used in SPM-3 (Selektiver Pegelmesser)

SPM-3 (Wandel&Goltermann) is probably one of the most popular portable "Selektiver Pegelmesser/Selective Level Meter", it covers 0.5-612kHz and it is ideal for 136kHz measurements, it has built in chargeable batteries for portable operation (but they are most likely defective).
It is a matching signal generator - PS-3 - which may be controlled by SPM-3 for tracking on the same frequency.


RF detector.
This is based on techniques found in HP-651B, but increasing the current through 2N3866 (BFR96) to 20mA LA7MI discovered that the frequency limit increases, -3dB limit was measured to 70MHz, it was used in his broadband RF millivoltmeter, constructed in November 72 - covering 360Hz-36MHx with 0.7dB variations. I is of course an old design, but sometimes once may still find such circuits useful, it is several other devices which simplifies operation like NE/SA604/614, TDA1576 etc.

HP651B detector circuit


Another linear detector which isn't so critical for matched diodes, almost any smallsignal silicon diodes may work has been tested in the range 50-5000kHz,
but I am not quite certain about all the critical factors.

Level. -10dBm -20dBm
50kHz 2.0 0.608V
190 2.05 0.654V
300 2.05 0.657
490kc 2.05V 0.660

Measured with WG PS-3 and Fluke 8020B DVM

The detector may well be connected to a 1mA FSD 2kW meter


2-way divider (LA7MI)


6-way divider (LA7MI)

A practical application in a multi-frequency xtal oscillator

Power dividers: See 32. Weinheimer UKW Tagung, Scriptum der Vorträge 1987:
Leistungsteiler für VHF/SHF (DJ1EE):

1) Hybride Leistungstrafos, breitbandig KW-UHF
2) 3dB Richtkoppler (Koaxial bzw. Streifenleitungstechnik)
3) 3dB Branchkoppler (Planarer Streifenleitungstechnik)
4) Mehrwege- Wilkinsonteilr (Planare Streifenleitungstechnik)
Leistungsteiler/Addierer nach Wilkinson, Spannungsverstärkung der Einzelverstärker,
Breitband-Wilkinsonteiler (144-432MHz)

Booster amplifiers, see page-L2 or c14


It was a problem to find 2m signal generator with sufficient stability for ssb/cw 25 years ago, while an RF generator was easier to get. Here is my up-converter using SBL1/SRA1. The oscillator with 43.333Mhz xtal produces 130MHz to mix with 14MHz from the generator, to give a signal on 144MHz. The mixer has 7dB loss. To make it easy, I added attenuator on the input and output for total 20dB lower level than the level generator has. To convert from dBU (600 ohm reference level, 0.775V = 0dB) to dBm (50W) you must add an attenuator of 10*log (R1/R2), where R1=600W. And if the generator has 75ohm impedance you could use a minimum-loss-pad.

IR test receiver - to test IR transmit diodes, any IR diode will work, I used a transmit diode for
my receiver


LA8OJ used defective MI SHF power meter heads to make his own probes for NARDA microwattmeter using subminiature lamps

#2.22) LA7MI 360Hz-36MHz broadband level meter.

LA7MI linear broadband RF mV-meter (Circuit diagram Part 1)

Part 2. Linear RF mV-meter
Some "Pegelmessers" have a selection for broadband, this is useful for several measurements, but usually this is only available for instruments covering up to 600kHz. Here is shown a broadband level meter covering from audio to 36MHz, and with some inaccuracies further up.
Many IC's are available for RF detectors, but they are not linear. A linear instrument has apprx 20dB ranges, but 10dB covers most of the scale.
Believe the circuit diagram of the RF detector was found in HP651 signal generator, but it was improved with higher current.
The rectifier delivers 30V DC

#2.23) 'Selective' RF level meter

"Selective" RF voltmeter for HF (two ranges) built in 1982 (described in Radcom)

N2PK's Amateur Radio Projects Page

Vector Network Analyzer (VNA)

Nolan Lee's site for tube testers

Old instruments and tube testers, see http://www.one-electron.com/FC_TestEquipment.html


Last update: 2004.10.18