Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 [2] 3 Next   Go Down
  Print  
Author Topic: Input Boost for Vintage Frequency Counter  (Read 89653 times)
K8AXW
Member

Posts: 7005




Ignore
« Reply #15 on: August 12, 2013, 09:52:52 AM »

Martin:  Thanks for the link and info.  My first regen receiver many decades ago worked great.  It was tube type and was really the one thing that sparked my interest in ham radio.  However, the solid state regen RX I recently built is very disappointing.  This is why I am always looking for a better circuit.

Dieter: 
Löwenbräu had to lay off a shift of workers when I left Germany in 1956!   Roll Eyes

I visited the Hofbräuhaus only once.  Damned if THAT wasn't an experience! 
Logged

A Pessimist is Never Disappointed!
KB1WSY
Member

Posts: 1306




Ignore
« Reply #16 on: August 12, 2013, 10:25:50 AM »

The counter uses a 1 MHz xtal and divides this frequency to get the time base. All those rectangular signals can cause problems if not fully shielded. So that is probably what you hear.

After reading your post, for troubleshooting, I encased the frequency meter inside a closed aluminum box, and grounded the box. This didn't eliminate the audio tone but reduced its volume by about half. In future I am planning to dismantle the Heathkit frequency counter (it is not a rare item!) and re-mount the innards inside a custom "station console" to match my other equipment and serve as the "frequency dial" for all of my "vintage homebrew" efforts (saves me the effort of having to calibrate mechanical dials). If I pay extra-careful attention to shielding, this may deal with most of the problem with that audio tone I guess. Perhaps I can add shielding around certain components on the PCB. Mind you, for a solid state device, this meter runs really hot (the top of the steel case is hot, not just warm) so I'd better be careful with ventilation too.

Another advantage of repurposing the frequency meter is that I can add the preamp stage inside the box instead of putting it at the receiver end. Or at least experiment with that idea: it may work better, or not, we'll see.

The flickering readout could come from either a to strong or to weak signal. I already had both cases.

I connected an attenuator potentiometer between the radio and the meter and this made things *worse* on 20m, not better. In fact it was unable to lock into the frequency at all, the moment you put in a few ohms of atten -- so it looks like more amplification is needed. Rats. When I finished that little preamp, I trimmed down the perfboard to leave only a small border around the edge. No room to adde another stage! Another day, another homebrew lesson learned.

There is an internal Sensitivity pot inside the frequency counter, which would have been adjusted by the previous owner. I was told he was an electronics technician and the item is in great shape, so I am a bit reluctant to mess with the calibration. But maybe I'll try that next, after first carefully noting the "start" position before I change anything.

Thanks for your suggestions/advice, very useful.

73 de Martin, KB1WSY
Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #17 on: August 12, 2013, 01:44:34 PM »

So I recalibrated the sensitivity on the Heathkit frequency counter, as described in the manual. It made no difference -- or rather, the counter is now slightly *less* sensitive than it was before and I am unable to get any usable readout from my regenerative receiver on 20m at all -- instead of skipping around + or - 50 KHz it now displays random stuff anywhere between 9 MHz and 15 MHz. It still works perfectly when I set the receiver to 40m or 80m though, when used with the recently built preamp. I think I will leave the counter calibration alone: at least I now know that it has the proper "factory calibration settings" (if you can ever say that about something that was built as a kit in the first place).

So I definitely need more amplification, at least on 20m and perhaps other bands (haven't tested them all). Not sure whether the preamp circuit I built can be boosted. One thought: the output from the voltage regulator is 5V but perhaps this is only needed because of the various CMOS/ICs in the NorCal project (which is for a complete frequency counter; I built only the preamp part of it). The JFET I used (model J310) has a 25V rating and the MPSH10 follower transistor has a similar rating. What about if I apply the full battery 9V, bypassing the regulator circuit? Do y'all think I will fry something? Here's the link to the schematic again:

http://tinyurl.com/lkkzw2n.

Yeah, I know, elementary questions. I'm still a beginner, especially as concerns solid state circuits. The only reason I included the voltage regulator in the first place is because the circuit diagram for the NorCal frequency counter's preamp section specifies that the power should be "VCC" i.e. connected to the regulator. There's no further info in the construction manual, except that it does say that the CMOS components in the frequency counter need voltage limited to 5V.

73 de Martin, KB1WSY
« Last Edit: August 12, 2013, 01:48:29 PM by KB1WSY » Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #18 on: August 12, 2013, 02:39:36 PM »

My first regen receiver many decades ago worked great.  It was tube type and was really the one thing that sparked my interest in ham radio.

Sometime in the future I'm thinking of building this one: http://tinyurl.com/lgqapkz.

.....the solid state regen RX I recently built is very disappointing.  This is why I am always looking for a better circuit.<<

What design was that? (So I will know what to avoid!)

One of the modern gurus of regenerative receivers is Charles Kitchin, N1TEV. He's published several articles including this one: http://www.arrl.org/files/file/Technology/tis/info/pdf/9811qex026.pdf.

I haven't tried to build any of his designs because of my current focus on "1950s/60s vintage homebrew" but when I finally get over that obsession I might have a go at building a "Kitchin design." One of his contentions is that the early pioneers of regen radio in the 1920s had a lot of good ideas that were later forgotten. One of the ones Kitchin has resurrected is that he advocates a variable-capacitor circuit for controlling regeneration, replacing the usual potentiometer circuit.

BTW I would *not* advocate building the 1968 "3-transistor receiver for the beginner" that I built. I had to make quite a few changes to get it to work on anything other than 40m and several people have said it is an unusual design. I like it but that's because "it's mine" and it will form part of my complete "Novice station" -- but nowadays there are better designs out there.

73 de Martin, KB1WSY
« Last Edit: August 12, 2013, 03:05:29 PM by KB1WSY » Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #19 on: August 12, 2013, 03:34:26 PM »

After recalibrating the frequency meter, that 400 Hz audio tone on the receiver has gone away, at least on 80m and 40m. Excellent! The Heathkit calibration procedure involves progressively turning *down* the level of the 1 MHz crystal timebase signal until it is just adequate for the meter to operate. As the timebase signal is reduced, the sensitivity control is increased in tandem, until you get the "threshold" timebase level below which the meter doesn't work. (Quite fiddly. Took me about an hour.)

Looks like the original level of the 1 MHz signal was too high and (as KA4POL pointed out) this was creating spurious signals.

Sorry, yet another post ... it was too late to edit the earlier one.

73 de Martin, KB1WSY
« Last Edit: August 12, 2013, 03:55:21 PM by KB1WSY » Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #20 on: August 13, 2013, 04:27:53 PM »

Problem solved! I had enough components left over so I built a new version of the NorCal preamp circuit, but this time I left out the entire voltage regulator circuit. It is now working great with my regenerative receiver, with the Nixie tube counter reading all bands from 160m through 10m. Pictures:

The new preamp: http://tinyurl.com/l4tqudd.

Twenty meters: http://tinyurl.com/k4xp4xj.

Ten meters: http://tinyurl.com/keene7c.

I still need to eliminate the 400 Hz audio tone, which is audible (even after recalibration of the frequency meter's Sensitivity pot and dialing down the level of the 1 MHz timebase as far as possible) on all bands 20m and above -- not very loud, but enough to hamper weak signal reception. I believe this problem is eminently soluble. My evidence is: all I have to do to eliminate the audio tone is grasp the regen receiver's cabinet with my hands on both sides. So it's a shielding/grounding issue of some kind, not rocket science.

I ran the preamp for a couple of hours and it hasn't gone up in smoke yet, so I guess there is no problem with nearly doubling the supply voltage from a regulated 5V to an unregulated 9V straight from the battery.

73 de Martin, KB1WSY
« Last Edit: August 13, 2013, 04:43:17 PM by KB1WSY » Logged
JAHAM2BE
Member

Posts: 316


WWW

Ignore
« Reply #21 on: August 13, 2013, 08:50:21 PM »

I haven't tried to build any of his designs because of my current focus on "1950s/60s vintage homebrew" but when I finally get over that obsession I might have a go at building a "Kitchin design." One of his contentions is that the early pioneers of regen radio in the 1920s had a lot of good ideas that were later forgotten. One of the ones Kitchin has resurrected is that he advocates a variable-capacitor circuit for controlling regeneration, replacing the usual potentiometer circuit.

This is a tangent, but based on your interest in regenerative receivers, I decided it might be appropriate to post: recent investigations (late last year) have provided evidence that the actual operation of the variable capacitor to control regeneration (generally called a "throttle capacitor") does not work as is generally believed or implicitly understood. It has been stated in reputable texts on regenerative receivers that the throttle capacitor serves to control the RF current flowing through the tickler coil, thus controlling the feedback. Everybody seems to pass on this implicit knowledge without questioning it. But this explanation now appears not to be correct; another insightful analysis by the user vladn of TheRadioBoard (the same person who figured out how to equalize the regeneration threshold level across the entire tuning range) revealed that, in his words:
Quote from: vladn from TheRadioBoard
My conclusion is that a typical throttle capacitor regen control is acting predominantly by varying the amount of negative feedback applied via the Miller effect to the gate(grid) rather than varying the amount of the positive feedback via the tickler coil.

For details, see here: http://theradioboard.com/rb/viewtopic.php?t=4516
Logged

KB1WSY
Member

Posts: 1306




Ignore
« Reply #22 on: August 14, 2013, 09:09:18 AM »

I ran the preamp for a couple of hours and it hasn't gone up in smoke yet, so I guess there is no problem with nearly doubling the supply voltage from a regulated 5V to an unregulated 9V straight from the battery.

I spoke too soon. This morning it abruptly stopped working and the prime culprit is over-voltage. This happened a few minutes after I disconnected the preamp's own, separate, worn-down 9-volt battery (which was showing 7.8V on my meter) and connected the preamp straight to the receiver's own battery supply (which shows 9.2V on the meter). I don't have a transistor tester and don't know how to test transistors without one so I'm not sure which of the two devices is kaput, the JFET or the NPN follower.

I still have spare semiconductors so I guess the next thing to do is replace them, then add some kind of voltage-dropping device. To avoid guesswork with resistors I will experiment with a couple of 1N4004 rectifier diodes in series: 9.2 minus a voltage drop of (0.7 x 2) giving 7.8V.

For what it's worth, delving inside the regen set I also noticed that the main 500 uF power supply electrolytic had become disconnected. I have no idea when that happened. The negative axial lead had snapped off, right at the point where it connects to the shell of the capacitor; I have to say that the leads on modern components are sometimes very thin in relation to the bulk of the component itself. I managed to solder it back on.

This is all being done off the top of my head so if anyone has any contrary advice, it would be welcome. Perhaps I should never have removed the preamp's 5V voltage regulator in the first place and am therefore setting myself up for another future semiconductor failure if I use the 7.8V supply.

I am learning quite a lot in the process but after this is over, I'd like to spend a few days learning more about transistors. Trial and error is all very well, but I don't really understand what's going on in detail. I guess the best way to go is the combine theory-learning and practical building. I never thought I'd want to learn a lot about transistors but this is becoming intriguing. (Also I must admit that it is rather easier to muck about with low-voltage, cheap semiconductors than with tubes; even though I am looking forward to getting back to tubes!)

It has been stated in reputable texts on regenerative receivers that the throttle capacitor serves to control the RF current flowing through the tickler coil, thus controlling the feedback. Everybody seems to pass on this implicit knowledge without questioning it. But this explanation now appears not to be correct....

Interesting! New discoveries concerning regenerative receivers, nearly 100 years after the fact!

73 de Martin, KB1WSY
« Last Edit: August 14, 2013, 09:28:39 AM by KB1WSY » Logged
G3RZP
Member

Posts: 1041




Ignore
« Reply #23 on: August 14, 2013, 01:01:56 PM »

if it led to a change in Miller effect, it would lead to a change in tune frequency. For an explanation, go to the 1936 editions of Terman's 'Radio Engineering' - it's not in the later editions.
Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #24 on: August 14, 2013, 01:43:13 PM »

So I rebuilt the new-version preamp completely, replacing the two semiconductors in case one of them was blown. I added the two diodes to get the voltage drop. Unfortunately the result is the same. There is some amplification going on, but it is not enough to trigger reliable readings on the frequency counter. I messed around for hours, trying various things and checking every connection. I checked the voltages. Everything is "nominal." The receiver is working fine, the preamp is doing some amplification, but the frequency counter either shows a row of zeroes, or a random reading. Occasionally I can get a reading on 40m but it is unstable. The other bands are not showing a usable reading.

Yesterday it was working beautifully and earlier in this thread you can even see the pictures to prove it!

Taking some time off from radio building (as few hours at least!!!) ... when it gets this frustrating, a time-out is appropriate! Meanwhile maybe someone will have some suggestions. Looks like I may have to find another circuit. The JFET and follower circuit had just what I wanted: high input impedance, simple design ... but apparently, no cigar. Here's the schematic again (except that I removed the voltage regulator and substituted diodes to drop the voltage to about 7.6V): http://tinyurl.com/lkkzw2n.

73 de Martin, KB1WSY
Logged
JAHAM2BE
Member

Posts: 316


WWW

Ignore
« Reply #25 on: August 15, 2013, 07:47:11 AM »

Is it possible that your amplifier circuit is parasitically oscillating at VHF frequencies, which would prevent adequate amplification of your HF signals for the frequency counter?

I noticed from your photos that you used a piece of perfboard for the physical construction, which is generally not a recommended technique for HF construction. It would be better to use a solid copper ground plane, keeping inputs and outputs physically separated to try to ensure that the outputs don't couple back to the inputs (through stray capacitance or inductance) and cause unwanted oscillations. Additionally, there are techniques such as ferrite beads on the transistor leads and small-value resistors in series with the transistor to suppress parasitic oscillations. Here are a few links to read about such matters; search these pages for the terms "parasitic oscillation" or "ferrite":

http://www.qrp.pops.net/fm-sup.asp
http://www.qrp.pops.net/vhf-2013.asp
http://www.qrp.pops.net/jfet-bias-2006.asp

These suggestions may amount to no more than a shot in the dark, but perhaps by applying more careful construction techniques (ground plane construction; inputs separated from outputs; beads/resistors to suppress parasitic oscillations) you might get your once-working circuit working again.

I believe it may be possible to check for parasitic oscillation by monitoring the circuit's current draw and observing if it changes as you poke around the circuit with your finger. If it does change, then that is an indicator that the circuit is oscillating (the idea being that oscillation will be disrupted through the additional load introduced through touching). If the current does not change, then the circuit is likely stable and not oscillating. I've not tried this myself; caveat emptor.

Regarding your overall idea of using an RF amp to display the regen's frequency on a frequency counter, I also once tried to do something similar and had only intermittently working results, just like you. I suspect that my physical circuit layout was totally inadequate considering the gain involved (the circuit I used had several stages of amplification and was laid out haphazardly), but I never did get around to rebuilding the circuit and confirming if a better layout would have improved performance.

Just for your information, this is the circuit I used:

http://www.sentex.net/~mec1995/circ/Activ8/activ8.html


Of course, the regen signal enters the circuit where the antenna is indicated (the original circuit was intended as an active antenna amplifier for a high-impedance whip antenna).
Logged

KB1WSY
Member

Posts: 1306




Ignore
« Reply #26 on: August 16, 2013, 07:14:50 AM »

Regarding your overall idea of using an RF amp to display the regen's frequency on a frequency counter, I also once tried to do something similar and had only intermittently working results, just like you. I suspect that my physical circuit layout was totally inadequate considering the gain involved (the circuit I used had several stages of amplification and was laid out haphazardly), but I never did get around to rebuilding the circuit and confirming if a better layout would have improved performance.

Thank you very much for your detailed suggestions. It could be a while before I can get back to this. On the one hand I do feel I achieved a "proof of concept" but I suspect the rebuilding and troubleshooting will take a while.

73 de Martin, KB1WSY
Logged
KB1GMX
Member

Posts: 1729




Ignore
« Reply #27 on: August 16, 2013, 09:52:58 AM »

Martin  KB1WSY:

Looked at the circuit.  That is not a stable circuit with voltage in that at higher than normal voltage it may stop
operating due to bias out of range for amplification.  However at higher than spec voltage (9V) I cannot see the FET
failing and the bipolar should survive it too.  The Jfet is not stressed at 9V and the bipolar is easily able to handle 9V
and worst case 19ma that is easy for the parts to sustain.   

However the gate of the fet if it were to touch something that has a very high potential difference might be killed. 
One possible source for problems like that on the bench is one power supply grounded and the circuit under test
having a large static charge (or lightening) or a ground that isn't quite ground.  Note that earth ground if not
bonded to house ground may have significant potential difference between them due to normal or unusual
currents on the house grounding system. 

FYI tube circuits are always troublesome for solid state test gear as that input cap to the fet is a near short
until charged!  Its possible to hit the input of the fet with excessive voltage and kill it.  What is excessive?
for the J310 or MPF102 type fets about 25V!   Use of protective diodes wired as reverse biased from ground
to gate (cathode to gate) and from gate to Vcc rail (cathode to rail) will clamp the input spikes to acceptable
levels.  Suitable diodes are the common 1n4148, 1n914 and similar.

NOTE: that whip amp shown is nice but that fet must be protected as well. 


However to test the bipolar all you need is a ohm-meter of any sort.  The test is a verification that the base-emitter
and base collector junctions behave as diodes and that collector to emitter (base not connected) looks like an open circuit.

A Jfet is tested the same way.  Gate-Drain and Gate-Source should look like diodes (open one way low resistance the other).
The difference from bipolar is the Drain to source will exhibit a variable resistance that may change if you touch the floating gate.

Allison
Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #28 on: August 16, 2013, 05:07:32 PM »

FYI tube circuits are always troublesome for solid state test gear....

Thanks Allison, lots to chew over. Just a quick aside: my regen set is solid-state, not tube. The nearest thing to a tube in the entire chain is the Nixie tubes in the Heathkit frequency counter and they are cold-cathode so only marginally describable as "tubes." They look cool though.

73 de Martin, KB1WSY
Logged
KB1WSY
Member

Posts: 1306




Ignore
« Reply #29 on: August 17, 2013, 08:31:49 AM »

After some careful experiments I have determined that the preamps I built (3 of them, with identical circuit and perfboard layout except one of them has a 5V voltage regulator) are all working. The semiconductors are functioning normally, so my hunch that I'd fried some of them with over-voltage was wrong. This fits in with Allison's remarks.

(1) At a regulated supply voltage of 5V the preamp barely suffices to trigger the frequency counter, and only on 40m, and not reliably.
(2) At 7.6V or thereabouts, the preamp triggers the frequency counter on 40m and 80m, giving an accurate reading, but only sporadically; once in a blue moon it works on 20m too.
(3) At 9.2V the preamp is almost totally useless.

In any case this issue of the supply voltage is surely a red herring. What matters is to get a reliable result and I suspect that JAHAM2BE is on the right track concerning the parasitic oscillations, among other things. I will rebuild a preamp, this time using a copper ground plane. As for the layout, my original one was pretty tight but I can probably make it even tighter. I am not sure how I could have separated the input and the output more than I did -- they are already on opposite ends of the board.

I know nothing about this type of construction so while I'm at the electronics store, I'll buy a book about contruction techniques -- they have many ARRL publications including, I think, something suitable for my needs.

I suspect I also need a bit more gain: the evidence being that sometimes the counter just gives me a row of zeroes, rather than a flickering random number. The regenerative oscillations from the receiver are by definition weak, so it's apparently a challenge to build a preamp with sufficient gain while maintaining stability.

It was frustrating but it is on its way to becoming "fun" again. Troubleshooting is hard when there is more than one fault, which I think is probably the case here. I did have that one glorious moment when it worked perfectly for a couple of hours, so I *know* this problem can be solved!

73 de Martin, KB1WSY
Logged
Pages: Prev 1 [2] 3 Next   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!