EMRFD Message Archive 14572
Message Date From Subject 14572 2018-03-08 09:50:00 jwolczanski Winding pot-core inductors I intend to build a number of the audio filters described in Rick' s "High Performance Direct Conversion receiver" and bought a number of pot-core inductors (PC-2213-77). I've found these cores to be incomprehensibly fussy - the inductance changes if the two cores are slightly mis-aligned, or the mounting screw is tightened.
I was going nuts trying to hit a certain value (33mH) until I discovered this aspect.
Is there a trick to winding these?
I suppose I could turn this "fault" into a feature (a variable inductor!), but I'd much prefer to settle upon a repeatable method.
I also recall reading a paper about winding coils for audio (from Rick?), but can't find it.
14573 2018-03-08 13:14:02 Jim Davey Re: Winding pot-core inductors Jerry:Last year I wound many pot cores for some preamps for 136 and 185 kHz. I had good results using TDK cores from Digi-Key. Here are a few hints:1. Buy the PC-mounting hardware! This includes a compression clip to clamp the core halves together with proper pressure. Do not use nylon bolts2. Simply line up the outside edges of the core halves with your fingers once the clamp is snapped in place.3. Use the proper bobbin for the core. I used Litz wire and put the layers down uniformly and tight. Most of my cores nearly filled the bobbin.4. Use a #2 pencil through the bobbin, with some Scotch tape wrapped to make a snug fit if necessary. This makes winding the bobbin very easy.5. When done, wrap a couple turns of tape around the winding (I used packaging tape cut in advance to the right width of the bobbin)6. When you drop the bobbin in the lower core, put a couple dabs of silicone sealant on the bottom face to prevent movement, since there is a little play in the clearances.7. I used gapped cores, so you cannot clamp them together in the center since there is a small air gap by design. The mounting hardware clamps put the pressure on the outside of core.8. You can adjust the inductance by pulling one turn off at a time from the taped assembly (you must remove the top half of the core, of course, to do this)With this, I was able to get to within a few percent of the design inductance. I used my AADE meter to check each core. The target inductance might require +/- a turn between cores wound at nominal 65 turns each.Jim K8RZ 14574 2018-03-08 16:41:52 jwolczanski Re: Winding pot-core inductors Thanks for the info Jim!
Unfortunately I'm now the proud owner of ten Amidon PC-2213-77 cores which seem to be different from the TDK models you use which have gapped cores with edge clamps. I was not able to find any such hardware on the Amidon site. These Amidon cores have the nylon bolts and that seems to be it for any mounting.
Oh well - I was pretty much doing all the right things, including the use of a #2 pencil. I even drilled a 1mm hole and stuck a wire in it for a "poor-man's turn counter".
The dab of silicon sealant sounds like a good idea. I also have an AADE meter. I can measure the inductance value of a core that I wound, lay it on the bench, pick it up an hour later,measure it, and find a slight shift in the inductance. At first I thought it was the meter, but I've got a few fixed inductors in my junk pile that prove otherwise.
The cores are brittle; I've already busted one. But I do notice the inductance seems to be relatively unchanged if I hand tighten the nylon hardware really tight.
If I do this again - I'll use the TDK cores. Do you have part numbers for the core assemblies and associated hardware?
14575 2018-03-08 18:10:48 Bill Carver Re: Winding pot-core inductors Ungapped cores have the highest inductance, of course. They saturate the easiest, too, although that's not an issue with R2 audio inductors. But with a gap in the middle the inductance is much less sensitive to the "fit" of the outside edge, and are generally more stable and the inductance more predictable. While you might get away with a mounting bolt in the center of an ungapped core, because presumeably the inside and outside surfaces meet at the same time, in practice clamping the outside edge is the way to go since you need stability.
14576 2018-03-08 18:36:05 Jim Davey Re: Winding pot-core inductors Hi Jerry:There is a bewildering range of cores available for different applications, both gapped and non-gapped. So without knowing your application, I am not recommending the cores I used, just pointing out some practical aspects that might help you out. I did a fairly wide product search and found out that very few cores are available unless you want to buy $500 worth, so after talking to several major vendors, I ended up at DigiKey because they were the only source for the material and diameter I could experiment with at MF, and even then, only a very limited range of material and size. I am on vacation now and don’t have access to my lab notebook nor my part numbers. I do remember that I used M33 ferrite for the 100-200 kHz range and measured Q’s of around 600. And in my application, I have tuning capacitors in parallel, so my inductance does not have to be dead on, and the inductance values I needed are a small fraction of what you are attempting.Google EPCOS-TDK cores and then DigiKey’s online catalog to narrow down what is actually available. FYI, there are cores that accept a tuning slug in the center stem which may work better for you.For the inductance range you mentioned in the first post, Rick used some shielded TOKO cores in the R2 and later receivers. I don’t believe those exact TOKO cores are still available, but Mouser does carry a similar line of cores from Fastron. Try this part number in Mouser’s online catalog : 434-02-333J.Jim Davey K8RZ 14577 2018-03-09 06:13:07 jwolczanski Re: Winding pot-core inductors Thanks Bill and Jim
Well, I've added a layer of masking tape to both sides of the bobbin and this morning, measuring the two cores that I wound last night, I see one was "stable" (no change)..and the other, a modest change. I'm stuck with these cores from Amidon - and this has certainly been an eye-opening lesson.
I've got all the time (and wire) in the world here - I might even try making a little masking tape gasket where the center parts touch...just to see if I can achieve some degree of repeatability. With powdered iron cores, I find the inductance formulas to be spot on - this is a little maddening.
I think I'm heading in the right direction, trying to stabilize the bobbin with respect to the pot core. I don't think this core style is ever going to be insensitive to the assembly, I just want to manage it. I shall report back!
14578 2018-03-09 07:36:06 Bill Carver Re: Winding pot-core inductors Given that the assembly is mechanically sensitive, a masking tape gasket (spacer?) might be an interesting experiment but not be stable in the long term since permeability is a very strong function of gap. Particularliy small gaps. Sounds like you have a lot of "stuff", perhaps you have some Kapton (polyimide) tape, which is used to wrap transformers. It's electrically and mechanically pretty stable stuff, undoubtedly better than masking tape. It's not terribly expensive, but I expect Home Depot doesn't carry it.
Good luck....but can tell you don't depend on luck. Great to see thoughtful construction is alive and well!
14579 2018-03-09 09:55:08 Brooke Clarke Re: Winding pot-core inductors Hi Jerry:
There are two fundamentally different types of pot cores. Some have precision mating surfaces and the highest Al
values. Others have a small gap at the center post and these often include an adjusting screw to allow tweaking the
inductance. The gap also allows higher level signals without saturating the core.
It might be possible to use some very thin )mylar?) tape to add a gap to cores that do not have a gap which would reduce
the Al and maybe make them less finicky.
Brooke Clarke, N6GCE
-------- Original Message --------
> Re: Winding pot-core inductors
14580 2018-03-09 11:46:12 Jim Strohm Re: Winding pot-core inductors Jim N6OTQIf you haven't ever touched kapton, it's probably worth a buck to have some to play with. Even if you just play with its heat resistance properties, it's pretty neat-o stuff.Folks, ping me off-list and if there's enough interest, I'll put together samples. Yes, I will mail internationally if I get enough to break even on postage and if you're not in a restricted nation. USA folks, know that I will likely demand a usurious profit of about ten cents to send kapton samples, my time NOT included :)Brooke,Use kapton tape. Google it. It's widely available on fleabay. And ... since I got some recently, I may consider sending some samples in exchange for greenstamps, if there's desire in the group. A small roll of kapton will cost you several bucks; I could probably send a decent-sized sample for a decades-expired IRC, an old piece of paper currency worth more or less one $USD1.00, or one greenback, or even for an SASE.
14581 2018-03-09 12:08:50 jwolczanski Re: Winding pot-core inductors Hi Brooke!
I took a piece of ordinary bond paper and made a little gasket that kept the two cores from mating in the area of the nylon hardware. The inductance dropped by a factor of about 10. Too much for my liking.
When I took the core apart, removed the paper - and reassembled the core, the inductance did not return to its "pre-paper-washer" state. I took the core apart, and wiped the mating surfaces with some acetone - no difference.
I then wiped down the inside of the two pot cores with acetone - and the inductance zoomed up close to a value I wanted. I then removed that narrow layer of masking tape that I had wound around the winding. The inductance went up a little more.
These cores need to be spotless - no tape residue allowed!
Never try to re-use those little narrow strips of tape used on the winding.
I now have two cores that are exactly what I need and they've been stable for most of the day. I've been handling them at regular intervals. Whew!
I wanted to make a few 88mH cores - I've used some of those famous 88mH toroids for various filters and diplexers. My 88mH toroid stash is down to three, and I thought it'd be kinda neat to roll my own. I've got one pot-core inductor that mysteriously went from ~90mH down to 76mH. So now I gotta apply my new-found experience to that one. So it doesn't look like I'm going into the 88mH inductor production business anytime soon..........