EMRFD Message Archive 14676

Message Date From Subject
14676 2018-04-30 17:57:02 Ken Chase Regen RX problem
Hi All

I am looking for help with a problem with a regen rx I built. The article I used is here


When I touch any of the controls, the frequency changes. I thought that as long as the variable capacitor's stator was at ground potential, this effect shouldn't happen.

73

Ken VA3ABN
14687 2018-05-01 04:44:34 Y. Hum Re: Regen RX problem
The question is which terminal of the cap is grounded, stator or rotator.


Ying 
VA3YH/VE2EO

14704 2018-05-01 16:14:38 Vern VanZ Re: Regen RX problem
Ken,

My understanding (And experience) has always been that grounding the rotor will minimize proximity effect.  There are other things that can give you this problem, such as inadequate shielding. But it sounds like you may want to reverse your rotor and stator connections. I haven't looked at the particular regent you referenced, so might not be possible...


73,
14705 2018-05-02 07:41:11 kb1gmx Re: Regen RX problem
Depending on the circuit its usually the rotor.  Reason is you come in contact with or very near the shaft.
IF the cap for some reason isi not in the circuit such that either are grounded then use a long fiber 
or other non conducting shaft to control it.

However some of the best Regen designs do many things besides that.

Mechanical stability nothing should move that isn't supposed to.

Metallic front panel that is grounded to the circuit ground.

Keeping all devices that may have a electric or magnetic field as far from the panel and user.
This may require more explanation.  In one regen I built I found the better result was both the 
tuning cap and the  regen cap were less affected if they were near the rear of the chassis 
and the control was via fiberglass rod.  This was true even if the part was mounted to the 
chassis and therefore grounded!   Same was true if the primary tuned circuit was away and 
in the clear.


Allison
14715 2018-05-02 21:49:54 Ken Chase Re: Regen RX problem
Hi Allison

Thank you for you input.

Yes the rotor is grounded, which I clarified earlier. The problem is not just the capacitor, even if I touch the volume or regen control, I get the problem.

I wonder if the problem could be due to a bad 2N2222A or 1N34A?

73

Ken VA3ABN

14717 2018-05-02 21:51:01 Raj Re: Regen RX problem
The stator may not be grounded properly or your variable cap may have a lug that needs to be soldered to ground.
Don't depend on your chassis or panel to ground it. Make the ground hardwired.

A picture will help!

Raj

At 29-04-18, you wrote:
 

Hi All

I am looking for help with a problem with a regen rx I built. The article I used is here

http://www.arrl.org/files/file/Technology/tis/info/pdf/0009061.pdf

When I touch any of the controls, the frequency changes. I thought that as long as the variable capacitor's stator was at ground potential, this effect shouldn't happen.

73

Ken VA3ABN
14718 2018-05-02 21:51:01 Ken Chase Re: Regen RX problem
Hi Vern

That might be difficult for me to do. I guess I'll have to float the whole capacitor, then ground the rotor.

I tried to determine what is happening by the photos in the PDF, but too difficult to tell.

73 Ken

14722 2018-05-02 21:51:30 Ken Chase Re: Regen RX problem
Hi Ying

The stator is grounded. Assuming the body of the cap is stator.

73 

Ken VA3ABN

14724 2018-05-02 23:57:14 Bill Carver Re: Regen RX problem
Ken, the problem isn't diode or transistor.  Looking at the ARRL paper, the tuning capacitor they used is mounted by two screws, and neither rotor or stator are grounded to the panel by the mounting hardware. In the picture of the regen there is a wiper at the front (knob end) of the capacitor, immediately behind the ceramic front plate  (the white rectangle), that's springy and presses against the shaft. This springy piece has a connecting strap....you can see it sticking up from the top-right corner of the ceramic. A wire is soldered to it, the wire going to the board. So the rotor is grounded to the circuit board with that wire.

Since the box is plastic, things aren't inherently grounded simply by being bolted together. In the snapshot of the regen there's a bolt at the right end of the front panel, and a wire going from that bolt to the regen circuit board. I'm sure its connected to the ground on that board.  So in the receiver picture the panel grounded by that wire and you need to have that wire, too. You need BOTH wires to the circuit board ground.

Both the capacitor rotor and the front panel are grounded by their respective wires, but they are only independent for the type of capacitor they used, which has its mounting screws insulated from the capacitor itself. If you used a capacitor which is mounted by a nut on the threaded shaft of the capacitor itself, then the rotor is connected to the metal panel. In this case RF current from the capacitor flows back to the circuit board on BOTH the wire from the wiper on the capacitor itself, and also on the wire that connects the front panel to the circuit board. The portion of capacitor current flowing on the panel itself will be affected by your hand. If you have that kind of inherently-grounded-by-mounting type of capacitor, it's mounting could be the source of instability you are experiencing.

So, if you have that kind of capacitor bolted to the panel, start by verifying the wiper of the capacitor has a wire back to the regen board, and the panel also has a ground wire. If so, then as TEST 1, I would suggest you try a "strap" instead of a wire from capacitor wiper to the regen board. A "strap" being a thin, wide "wire" instead of a round one. Hopefully you can find a piece of strap....or use a piece of bare house wire, and hammer it as flat as possible. Help any? Much?

As TEST 2 I'd temporarily REMOVE the wire from the panel to the PCB. Did that make it better? The idea is to only provide one path for the capacitor current to get back to the PC board, and eliminate RF current that might be flowing along the panel, and back to the circuit board, on the panel....which you are touching.

If it's lots better we know we're on the right track. But if it's still not good enough, my next idea is more difficult. I'm suggesting insulating the capacitor shaft from the panel, just like the two-screw mounting capacitor shown in the ARRL picture. This only applies if you have a capacitor which is grounded to the panel by the mounting nut on the bushing. Assuming that's your capacitor type, get an insulated "shoulder washer" and flat washer the size of the capacitor bushing. The panel hole will have to be increased to the diameter of the shoulder. For the common 3/8" diameter bushing the shoulder is usually 1/2". Remount the capacitor in the enlarged hole using the shoulder washer: now the capacitor shaft is not grounded by the panel, but only by the wire from the wiper back to the PCB. Now the return current from the capacitor doesn't flow along the panel.  Reconnect the panel wire temporarily removed in TEST 2. ....you've duplicated what is accomplished by the two-screw mounted capacitor they show in the picture.

I suspect every OT out there has built a regen receiver or two, and I think most will admit that obtaining stability can be a tricky process. I would expect Allison, and perhaps others, to comment on whether this sounds like a reasonable approach.  If so, and you want to try it, I pawed through my junkbox and can send you insulating shoulder washer hardware for a 3/8" shaft if you don't have any.

Bill  W7AAZ


14725 2018-05-02 23:57:31 Ashhar Farhan Re: Regen RX problem
In most of the large variable caps, it is th rotor that is connected to the body of the capacitor rather than the stator. 
- f1

14726 2018-05-03 07:29:13 Vern VanZ Re: Regen RX problem
Hi Ken,

Regens can be great fun, but also great pains. Either way, they are definitely very educational. Once you have solved this problem and have had fun using it, I would strongly suggest "not stopping here". Modifications such as addition of a RF 'buffer' amp (with unity gain) between the antenna and the detector will improve performance, as well as greatly reduce interference to other nearby receiving equipment (regens can transmit!).

Another mod is to consider replacing the air core tube form L1 with a Toroid.  I did both of these mods (with great results) to a two stage regen design published in Oct 1944 QST.  

The Sept 2000 QST article you referenced, contains a  link ( bottom of page three) to a site that describes a regen which IMHO is a great next step...

73,
14727 2018-05-03 07:29:21 Bill Carver Re: Regen RX problem
Farhan is absolutely correct. Ken. The "rotor" is the set of moving plates connected to the shaft. The "stator" is the not-moving set of plates that have one or more places to solder wires to. Most (but not all) variable capacitors have the shaft and the rotor grounded to the panel by the the threaded bushing that the shaft goes through. The capacitor shown in the picture does not have the rotor grounded by its mounting screws, its only ground is a wire from the wiper against the shaft back to the circuit board.

If your capacitor is bolted to the panel with a large nut on the shaft bushing, and the there is that grounding wire from panel to board, now then the capacitor rotor has two wires for capacitor current to get back to the board. And one of them is down the front panel to that wire connection. Because the box is plastic, the panel isn't well grounded by mounting screws. This combination sorta makes the panel part of the circuit, and your hand it touching that part.

In the olden days that term "hand capacitance" was a common problem of regen and VFO builders. It sounds like that's what you're experiencing. Because the capacitor that was used in that picture is mounted by two machine screws to the panel, that mounting does NOT ground the shaft or its rotor to the chassis. In that respect it is not like most variable capacitors. But in a regen receiver, or a high stability VFO for that matter, having a wiper against the shaft and a separate wire from there to circuit ground can be crucial to stability of the VFO and eliminating "hand capacitance".

While I would not suggest it be the first thing to try, as I described in the earlier email, you can obtain the same isolation-from-panel for a tuning capacitor that is mounted by a threaded bushing with the insulating washers. But since that requires disassembly and enlarging the mounting hole (a one-way street!) I would suggest first checking the presence of the rotor grounding wire from its wiper, and try a separate ground wire to the panel like they show, and try it without the panel grounding wire.

Bill  W7AAZ


14731 2018-05-03 08:22:32 kb1gmx Re: Regen RX problem
>>>I wonder if the problem could be due to a bad 2N2222A or 1N34A?<<<

No, the radio works.


If anything its likely a grounding issue either in the radio construction or more 
likely the fact that if the chassis is not grounded at RF and it is part of the 
antenna system as a result. Most  simple regens are very sensitive to 
changes in the antenna system.

Its one of the characteristics of regenerative detectors.  If you isolate the regen
from the antenna with a amplifier it tends to get better (more stable).  At the 
extreme of building regens I went as far as a down converting 75M radio with 
a regen running at 455khz.  Since tuning and all was not at 455khz (fixed tuned) 
it was possible to have both stable regeneration and improved selectivity.  But 
we are now in the realm of simple superhet radios. 

 FYI it was more or less copied from the Mountaineer (back in the 50s) using 
1R5, 1U4, 3S4 tubes (battery was 2 C cells and eight 9V batteries (67V nominal).  
I later tried it with transistors and it worked well. The difference was the 
Mountaineer tuned the detector and used a crystal controlled converter.  
I used a fix tuned the detector and made the converter tune as I didn't have 
the oddball crystals.  

For those interested I'll put a copy of the Mountaineer in the files section.

Allison/KB1GMX

14786 2018-05-14 21:16:37 Ken Chase Re: Regen RX problem
Hi Bill

Thanks for ringing in.

The body(rotor) of the capacitor is grounded to the regen ground plane and is not touching the front panel.

The front panel is at a right angle to the regen PCB and is soldered together along the edge where they meet.

73 Ken

14787 2018-05-14 21:18:12 Ken Chase Re: Regen RX problem
BTW, this is the type of capacitor I am using. See attachment.

73 Ken

14790 2018-05-14 21:18:24 Ken Chase Re: Regen RX problem
Hi Vern

I didn't notice the link at the bottom. I guess that regen is next in line.

73 Ken

14803 2018-05-14 21:20:26 Ken Chase Re: Regen RX problem
Hi Bill

My capacitor is bolted(soldered) to the main PCB. The shaft sticks through a hole in the front panel. The front panel is also PCB and is soldered to the main PCB where the meet. The rotor is part of the body and is grounded.

73 Ken