The AM Forum
March 28, 2024, 03:41:04 PM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: VHF oscillator  (Read 9617 times)
0 Members and 1 Guest are viewing this topic.
w5rkl
Member

Offline Offline

Posts: 125


« on: March 19, 2013, 11:48:31 PM »

I've been working on an old Regency MR-10B VHF FM High Band tunable receiver. All stages are working except for the oscillator V2B (6U8). I have swapped the 6U8 with a known good tube but no change in the symptoms.

The IF is 10.7Mhz. I don't know whether the oscillator tunes above or below the receive signal. My first thought is the oscillator tunes above the receive signal which if that's correct then the oscillator would produce a 175.7Mhz signal with the dial set to 165Mhz.

So far the oscillator will oscillate with the dial set to 165Mhz and the oscillator's output frequency is 213Mhz. That is way too high for the 10.7Mhz IF and a 165Mhz receive signal. Adjusting the dial below 165Mhz causes the oscillator to stop oscillating. Tuning above 165Mhz the oscillator frequency increases but only a small amount, 224Mhz with the dial set to 174Mhz. The oscillator, when it oscillates, is stable, more so with the dial set closer to 170Mhz. The oscillator's grid voltage varies from -2.3VDC at 165Mhz dial setting to -5VDC at 174Mhz dial setting. The oscillator grid voltage drops to less than -1VDC when the oscillator stops oscillating which indicates the oscillator is not oscillating. The plate voltage on the oscillator is 182VDC when the oscillator is not oscillating but drops to 115VDC when the oscillator oscillates which, at 115VDC, is what the plate voltage on the oscillator should be.

I have adjusted C38 (A7) which is a 1 to 6pf piston cap but it only raises or lowers the 213Mhz signal. Adjusting C38 to lower the frequency below 213Mhz the oscillator stops oscillating.

Unfortunately I do not have an VHF signal generator that tunes 150Mhz and above to use as a signal source or substitution for the oscillator so all adjustments and frequency measurements are by "trial and error" with a 50Mhz to 2.3Ghz frequency counter.

I'm presently at a loss as to what is causing the oscillator to oscillate on the wrong frequency. I have checked all components against the part list in the Sams PHotofact MR-10 listing and they are all correct. All resistors are within tolerance. There are only 3 fixed caps, 2 micas and one ceramic tube caps so not much in the way of parts. The oscillator plate coil, L5, also is the correct coil with the proper number of turns meaning no mods to the coil appear to have been made. Any suggestions?

73
Mike
W5RKL


* Regency_MR-10-schem.jpg (1121.33 KB, 3272x2526 - viewed 542 times.)
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #1 on: March 20, 2013, 04:08:21 AM »

My first guess is that the oscillator should be oscillating below the receive frequency.  Could be one of the fixed caps may be bad, but that would be rare.

Check the tuning cap for a short between plates or even dust between the plates.

Fred
Logged
MikeKE0ZUinkcmo
Member

Offline Offline

Posts: 444



« Reply #2 on: March 20, 2013, 06:05:59 AM »

Not much there to go wrong.  I doubt there is anything wrong with the two tuning caps, since they both make the oscillator change frequency.   If there aren't any shorted turns on L-5, I'd replace C12, that's the only other frequency determining component in the oscillator.  After that the 500pF and 15pF coupling caps if need be.

I'd guess all the front end coils are self supporting with no forms, and if so, L5 should have the same "look" to it as the other front end coils and not show signs of having been "stretched out of shape", which would reduce its inductance and increase the oscillator frequency.  At VHF it doesn't take much to significantly change the frequency.

Logged

Mike KE0ZU

Bold Text and PICS are usually links

https://mikeharrison.smugmug.com/
n1uvi
Member

Offline Offline

Posts: 47



« Reply #3 on: March 20, 2013, 09:12:34 AM »

check that 190 volts at the 6X4
and check L5 in the oscillator circuit and R8 the 10 K resistor
if the voltage is jumping up that high when not oscillating
then the load of the tube is dropping out, check the tube socket over
real good for bad pins and carbon tracks
and how and where are you measuring the oscillator frequency
Logged
Rob K2CU
Contributing
Member
*
Offline Offline

Posts: 346


« Reply #4 on: March 20, 2013, 09:39:02 AM »

Can you describe the physical properties of the coil?  Number of turns, diameter, and length?

I presume you have verified that the tuning cap plates do not short while meshing.

Besides just replacing the three fixed capacitors, you could try heating the ends of the tubular ceramic with your soldering iron. If is the type I envision, the leads are wrapped around the ends.
Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #5 on: March 20, 2013, 11:03:18 AM »

L5 has 4 turns, about 5/16's long, and about 3/8" in diameter. I could be off on these dimensions difficult to measure. The L5 coil specs, turns, diameter, etc., are not listed on the part list, only the Regency part number.

The coil turns appear to be equally spaced and I haven't spread them apart. L5's wire gauge is pretty thick and it's not easy to spread the turns apart which is a good thing I guess, not easy to spread them apart by accident.

I forgot to mention, there is a "mystery" component that connects between the junction of R8, C9, C10, L5, and ground. This component has no markings, is round and about the size of a single layer RF choke body, about half an inch,or less, long, the diameter of a older 1 watt carbon resistor. The only possible type of component this is would be a capacitor. If it is a capacitor, without any markings and it's not shown on the schematic, I have no idea what the capacitance value is.

I have cleaned the variable cap's shaft to frame flexible connection plus tightened up the piston cap mountings below the chassis (a couple of them, there are 3, were wobbling). The piston caps are held tight against the chassis by pressure from the washer and screw. Rotating the washer while holding the screw steady eliminated the wobble in the caps. Unfortunately, that didn't change the symptoms.

The receiver RF amp coils are fixed mounted on the side of the variable capacitor, between the fixed plate tabs and cap frame with a variable piston cap in parallel with it (L1 and C36 along with L4 and C37) as shown on the schematic. L1 and L4 have not been touch, at least by me. The only work around those coils is to tighten the piston cap against the chassis.

I have checked for shorts between the plates as well as dust or corrosion tips that could cause the plates to short, found none. I cleaned the plates anyway, ensuring the plates were not bent out of shape, made no difference in the symptoms.

I thought one of the caps may have failed but the mica caps all appear to be physically okay. There is a dog bone cap off the oscillator's plate to ground, C12, 15pf, that also physically looks okay. There isn't much to the oscillator so not much to go wrong.

73
Mike
Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #6 on: March 20, 2013, 11:14:11 AM »

check that 190 volts at the 6X4
and check L5 in the oscillator circuit and R8 the 10 K resistor
if the voltage is jumping up that high when not oscillating
then the load of the tube is dropping out, check the tube socket over
real good for bad pins and carbon tracks
and how and where are you measuring the oscillator frequency

I measured the oscillator output frequency at the screen and plate of the mixer, V2A, both show the same frequency.

I agree, when the oscillator is not oscillating the plate voltage of V2B will be high, approximately 185VDC as measured at V2B's plate, but drop to 115VDC when the oscillator oscillates. I've monitored the oscillator frequency and the plate voltage on the oscillator when the oscillator stops and when it starts and the voltages do change as I indicated. I'm surprised the oscillator didn't fail when I connect the VTVM to the plate of V2B but it continued to oscillate, only a slight change in frequency occurred when I connected the VTVM probe to V2B's plate and the oscillator kept going but I expected that to happen.

I have thoroughly cleaned the tube socket and inserted the tube a number of times to "scrub" the socket and tube pins clean. The socket pins appear okay, no cold solder joints seen and the tube pins are snug in the socket. I've replace V2 with a known good tube but the symptoms didn't change so it doesn't appear the tube is at fault.

73
Mike
Logged
n1uvi
Member

Offline Offline

Posts: 47



« Reply #7 on: March 20, 2013, 12:30:27 PM »



I forgot to mention, there is a "mystery" component that connects between the junction of R8, C9, C10, L5, and ground. This component has no markings, is round and about the size of a single layer RF choke body, about half an inch,or less, long, the diameter of a older 1 watt carbon resistor. The only possible type of component this is would be a capacitor. If it is a capacitor, without any markings and it's not shown on the schematic, I have no idea what the capacitance value is.




there seems to be a misprint on the schematics
I see 2 C3's and no c9 or c10 but I think I see where you mean
the c3 and c19 must be c9 and c10
I would try removing  the mystery component
I assume L5 is a choke
and looks to me that the oscillator couples to the mixer through the b+ supply?
coming off the plate
don't want no mystery component there
Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #8 on: March 20, 2013, 01:34:19 PM »

Yes, there appears to be an error in schematic labeling. There appears to be 2 schematics with different component labeling for the oscillator grid capacitors. The capacitance values are the same in both schematics, only the component ID labeling is different.

One schematic shows the oscillator grid caps labeled as C3 (510pf) and C19 (15pf) while another schematic shows the same oscillator grid caps labeled as C9 (510pf) and C10 (15pf). None of the schematic shows that "mystery" component at the junction of the 2 grid caps, L5, and the 10K resistor, R8.

Both schematics appear to have the same "366.6" number in the lower right corner and the same PHOTOFACT data in the lower left corner, including the same date, 1957.

I'm going to experiment by tack soldering various cap in parallel with this mystery components just to see what difference/changes, if any, it makes.


73
Mike


Logged
n1uvi
Member

Offline Offline

Posts: 47



« Reply #9 on: March 20, 2013, 02:47:50 PM »

Try removing one leg of the component thereby taking it out
of circuit and see what happens
if it is a cap it should be a very small value bypass cap but then again it certainly does not even belong there
nor does any other value capacitor belong there
whatever this component is, its placement is across the oscillator output and
the oscillator power supply to ground
a capacitor added there will / should change the oscillator frequency

Logged
KU8L
Member

Offline Offline

Posts: 64


« Reply #10 on: March 20, 2013, 02:51:50 PM »

Remember, it is usually the simplest of the possible scenarios that is the issue.

It would be nice to know if the osc is supposed to be above or below...I agree that it is most likely low side not high side.

Also, if you are measuring on the plate side of the mixer, you are going to get all kinds of combinations...Both differences and both inputs.  Can you try to loose couple the counter to just the osc?  without detuning it?  

Without a scope it can be difficult to see what the counter is really locking onto.

FWIW

Curt
KU8L
Logged
n1uvi
Member

Offline Offline

Posts: 47



« Reply #11 on: March 20, 2013, 02:54:00 PM »

good point Curt
you will see all the sums and differences
Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #12 on: March 20, 2013, 07:54:10 PM »

I can lay the counter probe next to the 6U8 oscillator/mixer tube and display the same frequencies I see with the counter connected directly to the mixer screen or plate. The counter can easily see the L.O. signal before I place it next to the mixer tube or connect it to the mixer's screen or plate so the probe isn't loading down the L.O. I thought of that before I started measuring the oscillator with the counter in the beginning.

Yes, I agree, a mixer will have at least 1 or all 4 of the following signals appearing at it's output:

1. Local Oscillator frequency
2. Receive signal
3. Sum of receive and L.O. signal
4. Difference of receive and L.O. signal

I received a second but older version of this receiver today via UPS. The older version MR-10 sold in 1957 and is basically the same as the newer circuitry except for the cabinet, analog dial, and dial cord string. Excluding the dial and cabinet, place them both side by side and you can't tell the difference. Even below the chassis is the same, excluding any replaced components by previous owners.

The older version MR-10 L.O. works and the counter displays the L.O. in the older version before I place the counter probe along side the oscillator tube. For example, setting the tuning to the bottom of the dial, 152Mhz, the L.O. output is approximately 141.3Mhz. That indicates the L.O. is oscillating "below" the receive frequency which answers the question "above or below the receive signal". I was able to tune 162.55Mhz, the WX frequency, by manually rotating the variable capacitor (dial cord needs replaced, it's broken).

Now I have to figure out what's causing the newer version's oscillator to not work. I'll compare the dimension of L5 between the 2 receivers to see if a previous owner messed with L5 plus compare voltage and resistance measurements between the 2.

I really appreciate all the suggestions and comments you guys have provided. They have been most welcome and educating. I'll update the thread as I go along.

73
Mike


Logged
n1uvi
Member

Offline Offline

Posts: 47



« Reply #13 on: March 20, 2013, 08:42:01 PM »

Please do let us know what you find Mike
Id be curious to know more about that mystery component too
good luck

Scott
Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #14 on: March 20, 2013, 11:33:04 PM »

After restringing the dial cord in the MR-10 the dial appears appears to be off by 1Mhz on the top of the dial and about 2Mhz on the bottom of the dial. The Weather station at 162.55Mhz tunes in at approximately 161.2Mhz on the dial. The local city police tunes in at 154Mhz. I could adjust the oscillator so the dial is right but for now, I'll leave it alone and focus on the MR-10B L5.

L5 in the MR-10, the older version, has the same number of turns and the coil's diameter is the same but turns are not spread apart as much as the turns in the MR-10B. I'm going to squeeze L5 in the MR-10B to see how much the oscillator frequency is affected.

Mike

Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #15 on: March 21, 2013, 02:01:57 PM »


If there aren't any shorted turns on L-5, I'd replace C12, that's the only other frequency determining component in the oscillator. 



Success!!

Mike your suggestion was the right one, C12 (15pf), the old "Dog Bone" ceramic cylinder cap, off the plate of the oscillator tube to ground, was bad. Replaced it with a new 15pf Silver Mica. The oscillator no longer drops out throughout the tuning range. The receiver still needs some work, audio is a bit low and sensitivity appears to be a bit low, may be caused by out of tolerance carbon resistors.

I replaced the other two caps BEFORE replacing C12 but there was no change after each cap was replaced. I squeezed L5 turns together and still no change. C12 was the last cap to replace and that did the trick.

73
Mike
Logged
Rob K2CU
Contributing
Member
*
Offline Offline

Posts: 346


« Reply #16 on: March 21, 2013, 03:01:03 PM »


Besides just replacing the three fixed capacitors, you could try heating the ends of the tubular ceramic with your soldering iron. If is the type I envision, the leads are wrapped around the ends.

It is common for these caps to fail by solder connection of wrapped ends to metalized ceramic tube. Sometimes, reflowing the solder will perform a temporary fix.
Logged
w5rkl
Member

Offline Offline

Posts: 125


« Reply #17 on: March 21, 2013, 03:52:59 PM »

The dog bone cap was the last cap I replaced But, before I replace it, I clipped a 15pf cap in parallel with it and noticed a change in the oscillator's output frequency. I then removed the original C12 dog bone cap and replaced with a new 15pf mica cap, I have quite a few of them, and the test produced positive results.

Squeezing L5 turns closer together didn't change the symptoms but replacing C12, dog bone, cap did.

I suspect there's going to be some resistors, all carbon except for the 2400 4 watt sand resistor, that are out of tolerance. That's the next step, check and replaced any carbon resistor that is out of tolerance.

Again, thanks to everyone for all the comments and suggestions, there were most helpful. This has been, and still is, a great learning experience in oscillators.

73
Mike


Logged
MikeKE0ZUinkcmo
Member

Offline Offline

Posts: 444



« Reply #18 on: March 21, 2013, 07:52:33 PM »

Remember, now that you've squeezed L-5, you have to return it to it's original shape.   Set the dial to some mid-scale frequency, doesn't matter what that is, and adjust/un-squeeze L-5 so the oscillator frequency is again the required 10.7 MHz away from your dial frequency.  First though, adjust your Oscillator trimmer cap, I fergit the C-#, to it's mid-range position so next time you can do it the easy way. :<)
Logged

Mike KE0ZU

Bold Text and PICS are usually links

https://mikeharrison.smugmug.com/
w5rkl
Member

Offline Offline

Posts: 125


« Reply #19 on: March 22, 2013, 07:52:33 AM »

Mike, I spread the turns of L5 apart right after I replaced C12. Thanks again for your suggestions and comments.

73
Mike
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.043 seconds with 19 queries.