Friday, March 31, 2017

A (somewhat convoluted) means of locking a "binary" (2^n Hz) frequency to a 10 MHz reference

DDS (Direct Digital Synthesis) chips are common these days with small boards containing an Analog Devices AD9850 board being available on EvilBay for a cost lower than one is likely able to buy the chip by itself!  While these boards are quite neat, they do have a problem (or quirk) in that you are not likely to be able to generate the exact frequency that you want - at least if it is to be an exact integer of Hz.

Let us take as an example one of those ADS9850 DDS boards available on EvilBay.  These come equipped with a 125 MHz crystal oscillator that will likely be within 10-20 ppm or so, but let us assume that it is exactly 125 MHz.

Other than the 125 MHz clock and some output filtering, the AD9850 DDS chip has nearly everything else that one would need to generate an output from DC to around 60 MHz - the precise limit depending on filtering - and its frequency is set using a 32 bit "tuning word".  The combination of the 125 MHz clock and the 32 bit tuning word means that our frequency resolution is:
  • 125,000,000 / (232) = 125,000,000 / 4,294,967,296 = 0.02910383045673370361328125... Hz per step - approximately.
For most purposes around 1/34th of a Hz resolution would seem to be good enough - and it probably is - but what if you wanted to be able to generate frequencies that were exact multiples of 1 Hz steps for frequency comparison purposes or to be able to generate precise, standard frequencies like 1, 5, 10 MHz, etc. - or even a very precise 1 kHz tone?

The quick answer to this is to pick a clock frequency that is an exact "power of two" Hz, and the closest 2n multiple to 125 MHz is 227 or 134.217728... MHz - slightly beyond the ratings of the AD9850, but it is likely to work.  (Depending on the high frequency requirements, half of this frequency - 226 Hz, or 65.108864 MHz might be used instead:  Other frequencies that are 2n divided by an integer such as 2n/10 are usable, too as an example.)

What does this change in clock frequency gain for us, then?
  • 227 / 232 = 0.03125 Hz per step, which is exactly 1/32nd Hz.
In this way, very precise frequencies that are a multiple of 1 Hz (and a half-Hertz as well) could be produced.

(Where does one get a 134.217728 or 65.108864 MHz oscillator?  This would likely require a custom-made crystal/oscillator or it could be produced using another synthesizer such as an SI5351A that, itself, uses a VCXO as its reference.)
Locking the DDS synthesizer to a 10 MHz frequency reference

It would make sense that if you actually needed to be able to set your frequency to exact 1 Hz multiples that you would also need to precisely control the reference frequency as well - likely with a 10 MHz precise reference from a GPS Disciplined Oscillator (GPSDO), a Rubium frequency reference or something similar.  Unfortunately, 227Hz is an awkward number that doesn't easily relate to a 10 MHz reference.

The most obvious way to do this is to use a second DDS generator board (they are cheap enough!) clocked from the same 227Hz source with its output to exactly 10 MHz using a frequency word of 320,000,000d, comparing it to the local standard and applying frequency corrections to (e.g. shift slightly) the actual frequency of the "227Hz" oscillator.

There is a less-obvious way to do this as well, so here is an example using 224 Hz - which is 16.777216 MHz:

For the 10 MHz chain:
  • Take the 10 MHz output and divide it by 625 to yield 16.000 kHz
  • Multiply the 16.000 kHz by 32 to yield 512.000 kHz
  • Divide 512 kHz by 125 to yield 4096 Hz
For the  2n Hz chain:
  • Divide any 2n Hz frequency down to 4096 Hz as a basis of comparison.  For 224 Hz we would divide-by 4096 using a binary divider.
(Depending on one's requirements, the precise method could vary with other frequency combinations possible.  The frequency of 512kHz was used because it was well within the operational range of good, old-fashioned 4000 series CMOS circuitry.)

Why would anyone use this second method?  Back in the 1980s I built a DDS synthesizer that used a 224 Hz reference (16.777216 MHz) that used a 24-bit tuning word to provide precise 1 Hz steps, but I also needed to lock that same synthesizer to a high-quality 10 MHz TCXO.  While it would have been possible to have built another synthesizer, a 1980s solution to this problem meant that an entire synthesizer circuit (or most of it, anyway) consisting of more than a dozen chips - some of them rather expensive - would have have to be replicated to do this one thing.

This seemingly convoluted solution required required only 6 inexpensive chips - a combination of 74HC (or LS-TTL) and some 4000 series CMOS devices.  For example:
  • Dividing the 10 MHz reference by 625:  A 74HC40103 wired as a divide-by-125 followed by a 4017 counter wired as a divide-by-5 to yield 16 kHz.
  • The multiplication of 16 kHz by 32 to 512 kHz:  A 4046 PLL and a 4040 counter wired as a divide-by-32 to form a synthesizer.
  • Division of 512 kHz to 4096 Hz:  Another 40103 wired as a divide-by-125.
  • Division of 16.777216 MHz down to 4096 Hz:  A 74HC4040 counter dividing by 4096.
The final step to lock the two frequency sources together was to use the venerable 4046 phase detector, outputting the correction voltage to the 16.777216 MHz oscillator.

A comment about using the 4046 PLL chip:

It's worth noting that because the 4096 Hz output from the divide-by-125 from the 512kHz source is a pulse rather than a square wave so it is not possible to use the "XOR" phase detector (Phase detector 1) of the 40406, but rather the flip-flip detector (Phase detector 2).  The "problem" with the flip-flop detector is that when the two frequencies are close, instead of having a constant train of pulses being output that are either at the reference frequency or twice the reference frequency, one will get occasional, brief pulses as the output of one of the flip-flops occasionally drops out of its high-impedance mode.

The problem with is that these occur (more or less) randomly and comparatively rarely, meaning that they they are at a rather low frequency and can get through the loop filter, causing extra jitter on the locked frequency - the 16.777216 MHz oscillator in this case.  The "fix" for this is to slightly bias the output of the phase comparator toward V+ or ground with a high-value resistor (100k-4.7 Meg, depending on the application) which will "pull" the output constantly toward one rail, forcing the loop to be corrected constantly meaning that instead of the occasional, narrow pulse, there will always be a string of pulses at a "high-ish" frequency that can be removed by the loop filter.  With the rather low "loop gain" of this VXCO configuration, "jitter" caused by the multiplication synthesis and divisions really doesn't show up in the 224 Hz crystal oscillator being locked.

Even though the 4046 CMOS with its built-in oscillator and phase/frequency detector is a bit "jittery", remember that this is being applied with a slow loop filter to a VCXO circuit with very low overall loop gain so this jitter averages out nicely.

(There are a number of modern phase/frequency detector chips that do not have this "jitter" problem.)
With the main 16.777216 MHz reference being a VCXO (Voltage-Controlled Crystal Oscillator) the above scheme worked very well, locking to the 10 MHz reference in a second or two.  Back in the 1980s the most accurate frequency reference that I had was a collection of OCXOs (Oven-Controlled Crystal Oscillators) and TCXOs (Temperature-Controlled Crystal Oscillators) with the 10 MHz units being easily referenced to the off-air signal from WWV to provide both an accuracy and stability of around one part in 107 or better.  Because, in our example, we are starting out at a much higher frequency (e.g. 134-ish MHz) we might divide this down to 4096 Hz - or whatever frequency our implementation required - using a combination of 74F or 74Axx logic and a (74HC)4040 counter.

(If our 134-ish MHz clock were produced using an SI5351A synthesizer, the PLL corrections in this scheme would be applied to its clock, which typically operates at around 27 MHz.)

Nowadays, with GPSDOs and second-hand rubidium references being affordable, the accuracy and stability can be improved by several orders of magnitude beyond that of the TCXO that I used those many years ago.

Having said all of this the question must be asked:  Is any of this still useful?

You never know!


This page stolen from

Tuesday, March 21, 2017

The 1J37B as a replacement for a 1L6?

The rarity of the 1L6:

Owners of the classic Zenith Transoceanic radios from the early-mid 50's will probably be aware of the pain involved if they have to buy a 1L6 tube (or "valve") for their beloved radio:  A "good" 1L6 - seemingly the tube that goes bad most often in these radios - can fetch up to $60 today, a significant fraction of what one might have paid for a second-hand radio to restore.

Figure 1:
The original 1L6 - a "not-too-common" tube even during its
heyday.  A "good" one like this is even rarer today!
Click on the image for a larger version.
One of the problems with the 1L6 is that there really aren't any good substitutes since this 30-ish MHz rated tube, a hexode  (a.k.a. "pentagrid converter") wasn't  commonly used in the first place, finding near-exclusive use in higher-priced battery-powered shortwave radios.  One of the few (almost) direct plug-ins that exists is the 1U6, which is apparently rarer than the 1L6 and requires some slight circuit modifications to accommodate its lower (25 mA) filament current.

There are other tubes that will plug in, but these simply don't work on the higher shortwave bands (e.g. the 1R5, intended for AM broadcast band battery portables and not for the higher shortwave frequencies) or, in the case of the European 1AC6, requires a bit of modification and has issues with radio alignment.  There is, of course, the electrically equivalent and comparatively easy-to-find 1LA6, but it's in a completely different form factor (e.g. a loctal tube rather than a 7-pin miniature) and requires either an adapter or a different tube socket.  Finally, there are the solid-state replacement options which are roughly comparable to the cost of a known-good 1L6 and while some are reported to work fairly well, they definitely lack that "tube" aura.

What now?

One of the sticking points is that the 1L6 serves both as the local oscillator and frequency converter:  One of the internal grids of this hexode is used as sort of the "plate" of the oscillator while a grid closer to the anode takes the signal from the RF amplifier stage and modulates the electron stream to mix it with the local oscillator to produce the 455 kHz IF - and it does this all with a filament that consumes just 50 milliamps at about 1.25-1.4 volts.  Without significant rewiring, this kind of rules out the use of pretty much any tube other than one that takes just 50 milliamps at 1.4 volts for its filament!

Having established that there really aren't any other 7-pin miniature tubes that are "close enough", what about broadening the scope to include something entirely different?

Figure 2:
The Russian 1Ж37Б "rod" pentode.  Approximately the same diameter
as a ball-point pen, it's overall length, minus leads, is about that
of a 7-pin miniature tube.  This specimen bears an early 1987 date
of manufacture.
Click on the image for a larger version.
This thought came to me at about the time I was first experimenting with some Russian Rod tubes as described in my December 31, 2016 posting, "A simple push-pull amplifier using Russian Rod tubes and power transformers" - link.

While that article discusses the use of a 1Ж18Б (usually translated to "1J18B" or "1Zh18B") pentode, there is another member of that family, the 1Ж37Б (a.k.a. 1J37B or 1Zh37B) that is also a pentode rated for operation to at least 60 MHz.  One property in its favor is that its filament voltage and current are "pretty close" to that of the 1L6:  Anything between 0.9 and 1.4 volts will work and the rated filament current is around 57 milliamps - a tad higher than the 1L6, but something that we can probably live with.

Doing a quick finger-count of the number of elements of a pentode and comparing that with the number of tube elements that one would need to simulate a 1L6 hexode immediately reveals a problem:  How would one use a pentode as a pentagrid converter when we are an element short?

The 1J37B to the rescue?

As it turns out, the 1J37B is a unique animal:  As a result of its construction using metal rods to form and modulate sheets of electrons rather having the grid-like structures of "conventional" tubes, it actually has TWO "first" grids that are pretty much identical - a construct that is often likened to that of a dual-gate MOSFET - but it is more likely to be akin two two FETs in parallel.
Figure 3:
The bottom-view pin-out and the internal diagram of the 1Ж37Б pentode.
Following the original nomenclature, the "grids" are referenced using the "C" designation - somehow appropriate even in English since this tube does not use "grid" structures at all, but control rods to alter the trajectory of sheets of electrons from the cathode.  As noted in the text there are two "first grids" that operate identically and (in theory) may be used separately, interchangeably or even tied together as a single "grid" for higher transconductance.  Because these tubes manipulate sheets of electrons, they are quite sensitive to magnetic fields!
Click on the image for a larger version.


The internal mechanical layout of the 1J37B is also quite interesting in that it is essentially two tubes in parallel, sharing the same cathode, screen "grid" and suppressor "grid" and plate connections.  In the middle, the identical sheets of electrons from the cathode go in two directions, each controlled by its very own "C1" control rod (e.g. C1' and C1").  Beyond C1' and C1", the structures of the screen, suppressor and plate elements are physically mirrored and connected together.

In comparing the specifications of the 1L6 and the 1J37B, the important specifications  (e.g. transconductance, capacitance, filament voltage and current) weren't terribly different.  Some of the voltage ratings for the 1J37B - particularly that of the screen, rated for 60 volts maximum - are below that which one would see when used as a 1L6, but those may be dealt with later.

What if we could use one of these two "first grids" and the "screen grid" as the basis of the local oscillator section and simply apply the input signal to be amplified and converted to the other "first grid"?  Because it was more like two tubes in parallel than one tube with multiple control grids I wondered if there was enough isolation to allow both oscillating and signal mixing functions to occur simultaneously.  I was a bit skeptical of this idea, even though I was the one that thought of it (as far as I know.)

I decided to try it.

Making the base

Figure 4:
Using masking tape, a "form" is made to set the shape and position of the
pins the pieces of 18AWG wire poking through two layers of masking
tape to protect the socket.  After dripping in the epoxy, the pins were
moved about to make sure that they were completely surrounded by
Click on the image for a larger version.

Rather than mess with the Zenith TransOceanic for the first attempt at this, a friend of mine (Glen, WA7X) rummaged through his collection of old radios and produced an old Motorola battery/AC radio that used 1 volt tubes - including the 1R5 which is (sort of) "pin compatible" with the 1L6.  Being a broadcast band radio I figured that if the concept was usable at all, the simple, nearly foolproof low-frequency circuits of such a radio would be the place to try it first:  If it worked there, there may be some hope that it would work in the ZTO.

I needed to make a fake tube base, but not having a dud 7-pin miniature tube immediately at hand - and remembering from my past how difficult it is to solder to the "bloody stumps" of the dumel-like wires on the carcass of a deceased tube's base - I set about making one.  I first covered the 7 pin socket in the radio with two layers of masking tape and then poked through this tape and into the socket seven lengths of bare, 17 or 18 AWG copper wire.  A ring of masking tape was then placed around the outside of these pins and some "5-minute" epoxy was dripped into the middle, carefully avoiding the upper portions of the copper "pins":  No doubt a small piece of plastic tubing or a taped-together ring of a sheet of plastic from a discarded "blister pack" would have made a nicer form than a floppy piece of masking tape, but it did the job.

Figure 5:
After the epoxy had started to set up, it was heated with a hot-air
rework gun set to about 212F/100C to speed up curing.  After
it had adequately set it was removed from the socket.  Here it is before
the wires were trimmed and tape and excess epoxy were removed.
Click on the image for a larger version.

Working the copper pins back and forth to make sure that they were surrounded with epoxy I allowed the requisite "5 minutes" for the "fast curing" epoxy to (somewhat) set. I then heated the contrivance with an SMD hot-air rework gun on its lowest heat (212F, 100C) for several minutes which immediately caused the epoxy to set hard enough to work once it had again cooled.

Carefully removing the "base" from the socket and peeling away some of the masking tape I trimmed the seven wires underneath to lengths comparable to that of typical tube pins and did similar to the top side.  I then had my 7-pin, solderable "tube base".

From this point on the wiring of the 1J37B to the base seemed pretty straightforward..

Wiring it up:

For the initial stab at replicating the function of a 1R5 the 1J37B was wired to the 7-pin base as follows:

1J37B Pin                       7 pin base connection for the 1R5
1 - Filament (-)                 [Pins 1]  Filament and suppressor grid
2 - "Grid" 1'  (C')              [Pin 4]  "Oscillator Grid" (G1)
3 - Grid 3 (suppressor)     [Pins 1]  Filament and suppressor grid
4 - Filament (+)                [Pin 7]  Filament
5 - "Grid" 1" (C")             [Pin 6]  "Signal Grid" (G4)
6 - "Grid" 2 (Screen)        [Pin 3]  "Oscillator plate/grid" (G2)
Plate wire (top)                 [Pin 2]  Plate

Or, put another way:

7 Pin base connection   for the 1R5    [1J37B Pin connection]
1 - Filament (-) and Suppressor Grid    [1 - Filament (-) and 3 - Suppressor Grid]
2 - Plate                                                 [Top plate wire]
3 - 1L6 "G2"                                         [6 - Screen Grid]
4 - Oscillator Grid (1L6 "G1")              [2 - Grid 1' - (C')]
5 - No connect (see text)                       N/A
6 - Signal Grid (1L6 "G4")                   [6 - Grid 1" - (C")]
7 - Filament (+)                                     [4 - Filament (+)]

Again, note that applying the word "grid" to the 1J37B, while descriptive of the function, is not accurate:  These "grids" operate more as control rods to deflect/direct the sheet of electrons being emitted from the cathode.

For replacing a 1R5:
Figure 6:
Right at home, the completed 7pin miniature tube base in the Motorola
"test" radio in the 1R5's position.
Click on the image for a larger version.

A bit of explanation about pins 1 and 5 is in order at this point.  For the 1L6, pin 5 connects to a pair of grids that surround the "Signal" grid (1L6 pin 6), but on the 1R5 the suppressor grid is internally connected to the "low" side of the filament using pins 1 and 5. Because the 1J37B is a pentode, the suppressor grid must be grounded which means that it would be connected to the filament low side as well.

Whoever made the radio could, in theory, use pin 1 and/or pin 5 for this connection and the mounting of components and there is no real way of knowing without inspecting the socket wiring.  Because of this it would be a good idea to connect both pins together when emulating a 1R5 unless you know for certain how this connection is made on the socket of the radio with which you are testing.

For replacing a 1L6:

When using a 1R5 as a "pinch hit" replacement for the 1L6 (it will probably work only up to about 10 MHz) the voltage applied to pin 5, which is nominally at about 85 volts, is effectively shorted to "ground".  In the Zenith TransOceanic H-500 there is a 68k resistor in series with that line which means that the current will be around 1 milliamp or so, reducing the potential on the "85 volt" line - also used on the screen of the RF amplifier - by 3-5 volts, an amount likely not high enough to be noticed.  If the intent is to never use this replacement in lieu of a 1R5 we would just leave pin 5 disconnected.

Trying it out as a 1R5:

For testing it out in the "1R5" configuration (e.g. 1R5 pins 1 and 5 connected together) in the Motorola radio I inserted a 10k resistor in series with the anode lead in order to monitor its current, but despite this inserted loss the faux 1R5 worked the first time.  The filament voltage across the 1J37B was 1.0-1.1 volts, well within its operational specifications and indicating that the other tube in series with it across its 3 volt "A" battery (a 1S5) was probably seeing an extra 0.25 volts or so on its filament.
Figure 7:
The first prototype - the 1Ж37Б (a.k.a. 1J37B) wired to the 7-pin miniature
base as a "1R5".  The two 10k parallel resistors and 0.01 capacitor
were inserted into the plate to monitor current.  For this prototype the leads,
insulated with PTFE spaghetti tubing, were intentionally left at their original
length to facilitate rewiring and inserting other components (resistors, capacitors,
etc.) in the circuit during testing.  For a "final" configuration the leads would
be shortened considerably which could reduce possible instability.
Click on the image for a larger version.

There was a minor problem, however:  At some frequencies the radio would start squealing - something that it did not do with the 1R5.  It is possible that there is a failing component in this radio somewhere, or it may also be that this faux 1R5 has enough extra gain to cause circuit instability, or a combination of both.  Despite this minor quirk, the results were encouraging as it is usually easier to dispose of extra gain than obtain it in the first place.

As a 1L6:

I then decided to try this faux 1R5 in my Zenith TransOceanic H500 with pins 1 and 5 still connected together.  While it seemed to work fine on the AM broadcast band, the radio got increasingly deaf with each higher band.  A quick peek with a spectrum analyzer on a service monitor showed that the oscillator was working on all bands, but it was always a low in frequency, causing mis-tracking of the RF filtering with the error increasing as one went up, being low by about 600 kHz on the highest (16 meter) band.

There was another problem:  On 19 meters the radio started to become unstable, behaving like a regenerative receiver on the verge of oscillation and on 16 meters there was just solid hash, indicative of instability - likely because of excess gain.  Referring back to the 1J37B specifications, I'd noted before that the noted maximum indicated screen voltage was on the order of 60 volts - but nearly 90 volts was being applied in the TransOceanic.  Because of the rather low current pulled by the screen grid (being used as the "plate" for the local oscillator) and the still-within-specs amount of plate current (around 3 milliamps).  I wasn't particularly worried about violating this voltage rating as there is no actual delicate "grid" that can be damaged, but it occurred to me that the gain could be reduced a bit by lowering the screen potential.  With a bit of experimentation I determined that a 33k resistor paralleled with a 1000pF capacitor in series with pin 3 of the 1L6 socket reduced the screen voltage to around 65 volts - still a bit above its specifications - but this change resulted in unconditionally stable operation.

(Comment:  This instability could probably have been reduced by using a less-haphazard wiring arrangement of the adapted tube, possibly along with some shielding.)

Disconnecting the now-unnecessary pin 5 and wielding an alignment tool I went to work re-tweaking the radio.  For all but the 16 meter band, the local oscillator adjustment was well within the range of the various coils and capacitors, but for 16 meters, removal of the local oscillator's slug only brought it to within about -400 kHz of where it should have been.

On the lower bands, particularly AM Broadcast, 2-4 MHz, 4-8 MHz and 31 meters, the radio's sensitivity was reasonably good - but not quite up to that of the 1L6 on 31 meters, but perfectly usable nonetheless.  For the higher bands, 25 and 19 meters, I could still hear a bit of ambient atmospheric noise and those radio stations for which propagation was extant, but like 31 meters, the receive sensitivity was still a bit low indicating the need for yet more tweaking.

More tweaking and testing:

I later did a bit more experimentation, adjusting bias and re-dressing the leads, but I could not affect the 16 meter tuning range significantly enough to bring it back into dial calibration, nor could I make a "dramatic" improvement in the high-band sensitivity.  If I'd replaced the core in the 16 meter oscillator coil with an aluminum or brass slug I may have been able to drag it up to frequency, but I didn't try it.


I did prepare another 1J37B tube and wired it in an identical manner to the first shown in the previous pictures (but with shorter leads) but interestingly it behaved remarkably different than the first:  It seemed to be much more prone to bouts of spurious oscillation (e.g. broadband noise) and fitful, intermittent local oscillator operation - a state not dramatically affected by swapping the two "first grids" C1' and C1".  Otherwise, the tube seemed to be behaving about the same in terms of DC current as the first.

What this told me is that my initial configuration of using the "screen grid" as the oscillator plate and applying the RF signal to be mixed to the other "first grid" may not be the best approach, as was my initial hunch - particularly in light of the fact that two seemingly identical tubes, both with fairly similar DC characteristics, seemed to behave radically different in this circuit - a strong indicator of a "non optimal" circuit topology that required one to work with "quirks" of each, individual tube!

In the future I may reconfigure the circuit a bit to see if configuring the tube in some sort of "Gammatron" configuration may yield better results - but that will have to wait until I get more free time...

* * *

Additional information about the 1J37B and the "Gammatron" mode of tube operation:

  • The 1J37B at the Radiomuseum - link (Includes discussions about operating the tube as a Gammatron.)
  •  Russian rod tubes at "Radicalvalves" - link (Information about the 1J37B and other "rod" tubes.)


This page stolen from