The AM Forum

I have an HRO-60 to go through. What caps to use is the question.

I used to use orange drops because they have a good reputation and seem more period appropriate. But somewhere along the line, I think it was an SX-62, I gave up and switched to yellow poly's because they fit in the crowded wiring. They are smaller and have axial leads. I have to admit, I don't like they way they look.  

Is there any technical argument for orange drops or is it just aesthetics?  And no, I'm not ready to do any capacitor re-stuffing.

Good question. I have measured the ESR and RF characterisitcs of both through 30 MHz and saw very little difference. Now the properties of the dielectric could be an issue in some applications and I cannot recall the dielectric type for Orange drop. So dielectric absorption in certain apps is a big deal and the poly I believe are better in those apps, like a PLL. Other opinions are certainly welcome.

Well there are always silk capacitors with oxygen free copper leads you could use to

"relieve the music's vibration energy"
and to "decrease the peak feeling sound at high compass and rough quality sound at middle compass"

http://www.thlaudio.com/elnaitme.htm

 :)  ???  ::)  ;D

Or I could buy Black Beauties prized by guitar amp builders.

Orange drop caps are high quality parts but they are not RF caps. I would not use them for low Z bypass above about 1MHz without a disc ceramic in parallel. High Z circuits they are no worse than a wax cap but far higher quality. They should out live you if used properly.

Tests myself and others have run on those to settle arguments show that the values used in that radio are series resonant in the mid HF range and are far from effective in the front end stages. Paper caps were even worse.

National started with all paper, then a mix of paper and mylar, then mylar and disc ceramic. This was over the 1952-68 production life.

Myself and others then made performance tests with the ugly OD's and discs from V-1 thru to the 2nd conversion but not IF stages.

The discs gave a noticable improvement on 10M of about a .5uV improvement for an AM SNR of 10dB. I also tested 10M but dont remember the difference but assume it was less and I didnt have a 6M AD coil.

Others have also reported that stock 60's with discs and out of tolerance resistors replaced (there are always many) were very sensitive even on 6M.

By the time I get thru with a typical 60 rebuild with selected tubes, a 6BY6 in both conversions and a 6GM6 mod for V-1 the 10M AM sensitivity is ~.25uV for a 10dB SNR.  It is one hot mutha and mine spends a lot of time on 10M.

Similar changes are made to the HRO-50, NC-183D and to the older NC-183 where a 6SB7Y is used for the mixer where it is a very decent 6M AM performer.

Carl

wooah. who buys this stuff? i dont understand.

checkout this resistor description:

I use Orange drops because I got hundreds and hundreds of them from ham fests new very cheap.  Later, I just purchased the ready made Orange Drop kits from Tubesandmore dot com.  they contain most of the values I would need and seemed to be a decent price.  

The one thing that alot of people complain about with the orange Drops is that the leads come out the bottom.  Most of the time, This means you have to bend the leads around to fit into the location where the tubular cap was.  The radios never look as clean with the Drops.  

I do not care for the Yellow tubular caps.  Some of them are polorized and being a rookie, I have to look over the schematic if its not obvious as to wich way the cap goes.  Then, I had two of them fail on me after a year. I also noticed that the Yellow tubular capacitors have very thin light gauge leads. This is a problem as I like to use needle noise and an iron to wiggle the lead THRU the lug on all connections. With the Yellow caps, This is very difficult as the lead just bends.

I have never had an orange Drop fail after install.

C

Orange drop caps are LARGE devices. Gee Whiz, of course their SRF is not going to be in the GHz range. So lets get calibrated and take some data. I think you all will be surprised. I walked down to the lab, fired up the VNA and measured a .047 uF/ 100 VDC orange cap. The data is attached along with a model of the cap to fit the data. Bottom line, the cap has a series R of ~ 5 ohms and a series inductance of ~ 35 nH. Not bad for a physically large part. Can you do better. Sure! The SRF by the way is ~ 10 MHz, however the series L is small enough that this part will function to 100 MHz.

Great info, Its to bad its for a 100Volt device.

C

Agreed, its not a valid test. In any case "work to xx MHz" is also not a valid argument.

Let me know what would be useful to measure. Its not difficult and I will be glad to combine into one plot and display. I went through this several months ago as I was contemplating what to use on another receiver restore project. I got hooked on this measurement process.... when for giggles and grins I decided to measure a 5 KV plate tuning capacitor at 14 MHz... just to see if it really was 15-150 pF. Can you imagine what the SRF of these monsters are??!! Alarming as it may sound... sometimes we are really tuning a variable L !

Carl, what do you use to replace leaky caps in an HRO-60 or 50?  Do you test disc ceramics?

I have to admit I assume paper caps are bad,  micas are a 50/50 proposition and ceramics are good. Is it time I revised my biases?

Not sure what you imply... "is not a valid test". I could bias the unit to 100 V, and re measure its Z vs. frequency. So, what is invalid? As far as work to xx MHz... yes, you would need to investigate if this actual Z vs. frequency is compatible with the remaining network, for example, a series R of 5 ohms is not negligible if I am matching to a 1 ohm device.

Orange drops used to have a black line on one end that indicated the outer foil.
This lead is supposed to go to the low impedance part of the ckt such as from cathode to ground....the line side goes to ground.
For coupling a plate to the grid of the next stage the line would connect to the plate.

There is a way to determine which lead is the outside foil....it is not always the same one.
Connect a scope to one lead and the ground lead to the other. There will be some noise pickup.
Then reverse the leads....the lead with the highest hum level is the outer foil.

I'm not an audio phool so don't any of youse jokers jump on me about this!  >:(  ::)
I didn't believe it at first but it's true and does make an improvement in noise.

They are not cheap like a disk but you get what you pay for they have a better tone to audio then others they say. But I think the frequncey so low or high we can't even hear them but you use film cap. looks like yellow shot gun shells.

Dave,
        I have always hooked them up that way! What sucks is like you said, they stopped marking the lead from the outside foil. that was the same reason that I didn't care for the yellow "tootsie roll" polys.

Orange drops are / were usually mylar dialectric, as were just about all of the "epoxy dipped" tubulars, and were immune to moisture / humidity issues, because they had nothing in them that could absorb it.

The problem with the early paper dialectric caps was the paper, as back in the day they used to bleach the paper with sulphuric acid. There was always trace amounts of the stuff left in the paper. Add any amount of moisture and the cap got an identity crisis. It didn't know whether it wanted to stay a capacitor, or change itself into a resistor.

One problem that I ran into with orange drops was that for a while they had tinned steel leads. (as did the tootsie rolls) Because the ODs were radial lead caps, you had to sometimes get "creative" in forming the leads to get everything where you wanted it, and I have had the leads come loose inside the epoxy while twisting and bending them. Even though the tootsie rolls had steel leads, this was much less of a problem due to the fact they were axial leaded caps.

This is a topic I have discussed ad nauseum at the Mid atlantic Antique Radio Club. And here is the main thing you ALWAYS should keep in mind when replacing caps in an old piece of equipment- - - - Whatever kind of cap you decide to use to replace an old waxed paper cap, (mylar, poly, disk, etc) is going to be 1000 times better than the one that you are replacing! ! ! ! ! ! and keeping in mind that disks and modern caps aren't marked for which lead is the outside foil, it really doesn't make much difference when using them for bypassing as the cap itself will keep itself at RF ground! ! ! !

I have repaired 100s of old radios and pieces of ham gear over the years. Unless you are replacing something that is application specific (HV, doorknob, silver-mica etc) It really doesn't for the most part matter what you use, so look for deals and dont pay bazillions of bux for fancy-assed caps that you dont really need! ! ! ! ! ! ! ! ! ! ! !  

Oh, yea, for what it's worth those white tubular caps are quite often early generation mylar dialectrics and usually dont go bad. You may want to remove 1 and bust it open to see what the dialectric is.

somebody said the ESR measures around 5 ohms. In a high Z tube circuit this is nothing but I want to roll on the ground when I see them used to bypass a low Z solid state final.
The performance limit of all these foil caps is the method the leads are attached to the foils. This usually adds some series inductance.
Sanders Associates made their own foil caps that were terminated in a way that reduced the series L to a very low inductance. I saw them blow away OJ caps when we were EMI testing a high power deice system years ago.

In older radios, like my 1939 airplane dial Sky Buddy, I have removed the defective original wax/paper caps, heated them up and removed the guts. Then I insert a small modern metalized film cap into the tube and resealed the ends with some of the saved wax from the old caps. I have used beeswax as well.

You can usually rebuild the old metal can caps like those in Command sets in a similar way.

Cosmetically perfect, and I have always liked the performance of metalized film caps. They are compact. Any technical magic to Orange drop or Black Beauty caps is strictly urban legend, IMO.

Correct! Yes, I measured 5 ohms series R on this 100 V OD, and that is awful for Solid State work. I guess the point of this thread at least from my perspective, if at all possible,  when these WISE tails rise up from the dungeons... let's measure. Again... In GOD we trust, all others bring data.  :)

That is all well and true for modern solid state stuff, but.................

This thread was originally started about an HRO-60 which is neither modern or solid state  ;D  ;D

I wonder if they even knew what ESR was back then  ???  ???

And to the original question posed, the only technical advantage to using them was better reliability and higher quality some years back, when the Asian offerings were questionable at best.

As for any other reason, it's been batted around here numerous times. Until recently, ODs were manufactured right here in the good ol' USA, so it was an opportunity to support a 'local' supplier. They are also great quality, offered in a large variety of sizes and voltages needed for old equipment. They still are made here, but on a limited basis now, mainly custom runs. The bulk are now made in China.

The biggest drawback is the radial lead arrangement, not well-suited for all environments. Though they do have long enough leads, this combined with the size of the higher voltage pieces makes them difficult to fit into some rigs. This does not lend them well to re-stuffing old caps, of course. And if you're old like Carl, bright colors make you dizzy.  ;D

Never been inside a HRO-60, but the timeframe is right. They'd probably be right at home, especially if you had some of the older Sprague-labeled units.

I usually use orange drops or disc ceramics, depending on their function, to replace old-fashioned wax-paper caps. I try to find new disc ceramics to replace failed ones, and good micas to replace failed micas.

I have boxes full of n.o.s. and salvaged capacitors of every description.  I usually just try to find something that is compatible in my junkbox that checks good, using a capacitor checker to test for leakage or else the brute-force method of DC power supply and mirco-ammeter (hoping the cap doesn't abruptly short out while under test), because I usually need it RIGHT THEN, not a few days later when one I ordered online comes in.  If I need it immediately and can't find a replacement, I try the local electronics store (not Radio Shack), but there is only about a 50-50 chance they'll carry it. I haven't had a great deal of problem with replaced caps crapping out, but that could explain some of the intermittents I seem to be plagued with.

Carbon composition resistors seem to be a greater problem than the old wax caps.  I'd say over 75% of the ones I check, whether residing in existing equipment or still n.o.s., have drifted in value, anywhere from 30% to over 100%. In non-critical parts of the circuit, I rarely see any improvement when replacing one of these with one of the proper value, so I just  leave them alone. In critical applications I may buy a new one if I can't find one that has not drifted.

Old mica caps bear watching, too.  Disc ceramics seem to hold their values, but I have had quite a few of these to short out (that's why I use two in series for the blocking cap between plate and mechanical filters in my 75A-4).

Like Bill, in the relatively rare event when I am interested in preserving the appearance of a piece of antique equipment, I re-stuff the caps, even to the point of re-using the original leads or same size wire (the newer caps tend to have smaller gauge wire, which not only detracts from the appearance; it is often too flimsy to hold the re-stuffed capacitor in place).

For replacing the old body-tip-dot resistors, if I want to maintain the original appearance and cannot find an exact replacement, I take the ohmmeter and look for one the proper resistance value, regardless of its nominal value.  I have found that even the drifted ones are fairly stable; maybe they have gone through most of their drift and settled down to their new value and will stay there. One could even re-paint the body, tip and dot to indicate the new nominal value, although  to date I haven't gone that far.

I tend to stay with carbon comps for anything that handles considerable power, since I have found the newer film resistors to be delicate, often opening up at a relatively slight overload or an accidental short with a meter test probe.

I replaced many of the by-pass caps in my 75A-4 with orange drops, and several of the little red mica coupling caps with newer micas.  It would be tremendously laborious to replace all the 1/2-watt carbon comp resistors, but the day may come when that's the only way to keep some of this older stuff going, assuming we can still find working replacement tubes by then.

Anyone know anything about the American Capacitor Corporation?  Looks like  they make caps in the US of A. and  not hand rolled, bees wax and paper sonic superiority stuff. The first distributor I checked will accept orders of $30 or more.  I may go out on a limb and buy some of their stuff  (if I can) just to avoid filling the HRO-60 with Chinese parts. Yes, I know this makes little difference in the big picture but at least I won't be thinking about China every time I turn the big tuning knob.

http://www.americancapacitor.com/

pictures

http://www.americancapacitor.com/case.htm

or maybe these guys

http://www.filmcapacitors.com/

I can't bring myself to re-stuff caps. Bill and Don are some kind of restoration heroes for doing it.
 

Metal oxides and are much more stable than carbon
comps. If you just replace caps and don't check the resistor values it will end
up biting you in the end.

Hammarlund switched over to disc ceramics at least as early as the HQ-140
for coupling and bypassing. 1kV discs are cheap and available.

I'm not sure ESR is a valid measurement for mylar or disc ceramics, since most
ESR meters are using 50 or 100 kHz for measuring ESR.  I'd expect their internal
reactive losses to be more of a concern, and that means making
impedance measurements in the HF ranges they'll be used for.

I'll restuff caps in radios that are early and rare enough to warrant the time and
attention.

All IMHO.

Pete

Especially with old National receivers!

W4AMV used a vector network analyzer from 0-100MHz to measure the ESR in his post above.   I've performed similar measurements with a HP3577A VNA on various capacitors, including ceramic disks.  The limiting factors are the lead inductance (which is really independent of the construction of the capacitor) as well as parasitic capacitance to ground.  In the latter case, tubular caps are much worse than ceramics because ceramics have a much smaller surface area, and the two capacitance "plates" are symmetrical, so they have common-mode parasitic capacitance to ground as opposed to tubulars which exhibit a much higher capacitance to ground from one side or the other due to the outside foil.

Hi. The test you performed was on a 100 Volt capacitor which would not be used in the HRO 60 we are talking about here.  In that radio the cap ratings on mine where 400 Volt or 600 volt.  Had you tested a new modern 600volt orange drop, the test would have been at least more interesting. I have a box of 100 volt and a box of 600 volt Orange drops and physicaly, They are very different. In fact, Nothing is the same other then the color.  There are also three main versions of the Orange Drop from my supplier.  The 715P, the 716P and the standard PS.  Various differences there are 5%, 10% and 5% with copper lead.  I use 715p and 716P most of the time.

I dont think it matters on the HRO either way, but I am paying close attention so I can learn more about caps...

C

Anyone know anything about the American Capacitor Corporation?  Looks like  they make caps in the US of A. and  not hand rolled, bees wax and paper sonic superiority stuff. The first distributor I checked will accept orders of $30 or more.  I may go out on a limb and buy some of their stuff  (if I can) just to avoid filling the HRO-60 with Chinese parts. Yes, I know this makes little difference in the big picture but at least I won't be thinking about China every time I turn the big tuning knob.

http://www.americancapacitor.com/

pictures

http://www.americancapacitor.com/case.htm

or maybe these guys

http://www.filmcapacitors.com/

I can't bring myself to re-stuff caps. Bill and Don are some kind of restoration heroes for doing it.


this ones I shot gun shells

I have a bunch of those caps. I've never had any problems with them. Note, anything with a mil number should be a good part that was built to a mil spec and test method.

Those Poly caps from American Capacitor are nice! ESR in milli-ohms at 100 Khz...

http://www.americancapacitor.com/vcover.htm

Of course we might find out that they too outsource from overseas...

I have used some cheap Polypropylene Caps from Mouser at HF in bypassing. On my Central Electronics 20A, at the buffer stage B+ bypass cap, there is high circulating current from the hi-Q L-C tank. Here the bypass ESR is critical and has a profound effect on stage gain and linearity. When using the Polypropylene cap (.005 uf) the gain went way up. I had to insert a 2 ohm resistor in series with it. That was too much, and I ended up with three 2 ohms in parallel. This suggests to me that the cheap metalized Poly caps from Mouser have an extremely low ESR, suitable for use in low impedance bypassing, even at RF up to 10 Mhz or higher.

Aren't orange drops available in multiple flavors? I mean polyester, polypropylene, etc.

   

Jim
WD5JKO

JN

Isn't ESR usually a spec that shows a series resistive loss in a capacitor at a particular
frequency, the same as the internal resistance for a (battery) cell? Effective Series Resistance...

If you're measuring parasitic capacitances and inductances, where does the "R"
component for ESR come from?  

I always understood ESR to be purely resistive component, with power loss.
Reactive components shouldn't introduce power (heat) loss.

A few searches for ESR yielded this:

In a non-electrolytic capacitor the metallic resistance of the leads and electrodes and losses in the dielectric comprise the ESR.

I meant to say "resistive losses" instead of reactive losses in my first post.

Pete

We've Got Yellow caps and Blue Resistors: (Sung to the tune: He's got tan shoes and pink shoe laces)

Tom, N0JMY, of Hayseed Hamfest is pursuing American made caps.  ESC/Westech does have distributors so maybe this will work out.

http://www.hayseedhamfest.com/
http://www.filmcapacitors.com/

They might be yellow, though.  I don't know.  Someone said bright colors make Carl dizzy. They make me dizzy too so I'm hoping for white, not yellow. What's with yellow? Who came down from the mountain and said from now on all capacitors will be yellow or orange?

Dave at justradios didn't seem as interested in chasing down US caps maybe because he's Canadian but he has started stocking new, not NOS, made in USA carbon comp resistors. I have a big stock of blue metal films and a pile of NOS carbon comps with the emphasis on Old but I'm thinking about these resistors. They aren't blue.  Maybe I'd trust them more than my ancient ones.

http://www.justradios.com/reskits.html

ASC caps are advertised as Made in USA on some cap site. They do have a facility in Nebraska. However, ASC stands for American Shizuki Corporation of Japan. Allied has them in their line card but the Allied website is down so I don't know just what they offer.

http://www.ascapacitor.com/

--------

... a polka dot vest and man oh man

I use discs on most of mine unless Im restuffing using those yellow and brown film caps and with 30's radios Im rarely pushing them above 20M. For audio stages I use whatever is handy and cost effective when values and voltages are high. Ive not fallen into the audiophool myth of piezo effect in discs at least at any power level even an early Super Pro can generate. Even my Scotts have discs in places.



Unless you use them above 20M or so the inductance is so low in say a .01/400V that you dont care how they go. It was a carry over from the old paper caps that actually used the metal foil as a shield between stages or between grid and plate. Some 30's radios will oscillate with modern caps but its simple enough to put 1/2W 10-20 Ohms right at the grid terminal and 100 Ohms at the screen pin with a disc bypass across both sides of the screen resistor, use carbon only, either kind. All it does is to brute force neutralize and oscillation suppress those old tubes with long internal leads.


The 400 and 600V ones most use in .047 and .1.  Simulate the actual lead lengths used in radios including the associated wiring. The radios are full of stray L and C that must be considered.
By not valid I was referring to the small physical size with 100V caps.


If that happens they are poor choices for that particular circuit.


Pure audiophoolery and there are several sites devoted to debunking them


HP and others used a film Black Beauty in the 50's and all the ones Ive tested in gear or NOS are still perfect. They have red lettering.



I like to be dazzled by young blonds and readheads, not bluehaired old cows ::)


Micas are going leaky left and right the past 10 years or so and getting worse, even NOS of the sealed dipped variety. Its a silver migration issue Ive been told by a ham chemist. Early dics were also a problem into the 60's. National got stuck real bad by CD with .1/50V in the first run of the HRO-500 and I replaced literally thousands while there as they were used everywhere. Later the .01's at 500V used in the NCL-2000 and elsewhere started going leaky or intermittent....they are all black discs. Ive also replaced a lot of tubular and dogbone ceramics, especially in Collins and some National. Be careful as some are used for TC and its pretty hard to find some TC value replacements.


I tend to do the same as Ive thousands of NOS and long lead used ones that have been staying well within tolerance. However Ive had to repaint many of the old dogbones using whatever I had that came within tolerance of what I needed; I figure after 60-80 years they are thru drifting.
Lately Ive been using the new Ohmite OX/OY family when a 1-2W carbon replacement is needed. They seem indestructible even in parasitic suppressors and as RF swamping resistors in older Alphas. etc. Available at Mouser of course.


Dont know anything about ACC but Ive been restuffing for at least 30 years and do them regularly for customers on high end ham and consumer sets as well as selected ones of my own. I know several who enjoy it so much they even do it to AA-5's that they own and a few do it to vintage TV's.


Their basic cap kits are all no name Chinese crap as you find at hamfests and radio shows. There have been many reports of failures over the forums and Ive stayed far away from them. They do have some HV 'lytics of 500V and up Ive been tempted to sample as they actually have brand names but Ive no experience with them. I tend to stick with what I know works.

OK, Im through for now having started at the top and worked down so excuse some duplication that others already addressed....Im to lazy to edit ;D

Carl

You are correct, but ESR is necessarily going to be a complex number - I think since the complex portion isn't really useful the measurement is generally written as a single-term impedance.   Note that W4AMV posted a Smith chart with his analysis noting the varying impedance (that's really what it is) at various frequencies.  As you know Q factor more commonly used in RF circuits as a measurement rather than ESR, but however you want to discuss it, 5 ohms in a ~10K-100K impedance tube circuit is a negligible factor.  The self-resonant frequency of the component, although the Q is very low, could be significant however.

Piezo effects of ceramic caps is real. I'm not sure an audio fool would notice the effect but you hit XR7 with a lightning pulse and the value dives during the pulse. I'm seen the value drop to 1/2 during a lightning pulse. It recovers if the dielectric isn't cracked.

If you got hit with a lightning pulse your own crack would be reacting too ;D ::)

Chapter 7 of this Op Amp Applications Guide has a very good discussion on the effects of real-world capacitors and resistors, including piezo-electric effect in ceramic caps, dielectric absorbtion, etc.  None of these is all that applicable to re-capping a 60 year old tube radio, but it's interesting to see how differences in capacitor construction can affect circuit design.

http://www.analog.com/library/analogDialogue/archives/39-05/op_amp_applications_handbook.html

TNX John, I collect this stuff.

Thanks for the link. Chart from the end of chapter 7:

Another interesting aspect of capacitor replacement is the values chosen.

Back in the days of crappy waxed paper caps when failure was very common. Many companies used 600v caps for some circuits because they were thought to be more "robust". If you looked at the circuit that they were working in, you find that they were running at 150v (or something like that). A real good example is : Why do you need a 400v rated cap in a cathode bypass application? ?

Also, many cap capacitance values were chosen strictly by price more than the circuit constants. After all, the actual cap tollerances ran anywhere between
+20-100%. Manufacturing tollerances were not all that accurate back then unless you bought the real high end or "special purpose" stuff.

I have given a handfull of talks at the Mid Atlantic Antique Radio Club (and others have discussed it ad nauseum) on recapping radios and selecting capacitor values for the application they will be working in. And to this day I still hear members discussung "where did you get those .0XX caps at".  If they remove something like a .05 @600v cap, they wouldn't even consider replacing it with a .047 or even a .022 in place of a .02  or using a 400v cap in place of a 600v cap when it is used for a screen bypass running at just over 100v. Geeeeeezzzzeeee......................GIVE ME A BREAK! ! !

Let's face it, a good a new .02 is still better than a bad or shorted .03..................

Oh yea, just FWIW, I usually keep a szhtload of 400v "epoxy drops" around (the small ones) in the more common sizes and use them for just about everything.

Every once in a while if you go threw Mouser or Digi-Key's catalogs you will find an odd value cap that they must have special ordered and got stuck with at a very cheap price. One time I bought 500 .0123s @400v for something stupid like 15 or 20 cents a piece. I dont remember the exact price, but it was real cheap, so I loaded up on them. Sounds like a .01 to me................ ;D  ;D

I buy a lot of my stock that way from local and fleabay surplus dealers.

For a bypass the beancounters approved the lowest value that would work in many cases and the difference between a .01 and .015, .05 and .068, .1 and .14, etc are totally meaningless. On top of it they are usually all industrial grade premium types, not no name China crap.

I scored several 500-ct boxes of super-dooper high quality polyprop .082/630V caps on eBay some years ago for about a half penny a cap.  I sold several boxes for $50 and still have some - but nobody wanted 'em on eBay apparently because of the odd value.  I use 'em to replace .1 and .05s.

Agree with the cathode bypasses.  I have scads of 200V caps that always go in that circuit instead of using 600 volters.

To add to the confusion level, one should consider that different dielectric materials result in different "temperature coefficients of capacitance" (read:  the capacitance is different when it's warmer)!

Admittedly, this is not often a big deal in typical bypass and coupling applications, but in some applications one must be on guard!

Thus, where a capacitor's actual capacitance is performance-critical, one should select capacitor types having low temperature drift.  This is generally more true of tube gear than solid-state gear, because of the difference in operating-temperature changes, and more true of any equipment subjected to greater temperature changes than encountered in the cushy environment of an ordinary hamshack.

My hazy memory suggests that mylar caps are not very temperature-stable at all, compared to some other types of caps.  Polycarbonate and polystyrene caps are much better in that regard.

The same hazy memory tells me that disk-ceramic caps are available in several types, with some having excellent temperature stability and some being downright lousy.  That is true of disk caps generally, not just the "temperature compensating" types that have specified temperature coefficients.

Tantalum caps generally have better temperature characteristics than electrolytics.

All of the above thoughts are from my pre-retirement memory of 20 years ago - time marches on - so does capacitor design - and one should carefully review current engineering literature.


-al hart
www.w8vr.org

For your viewing pleasure in case you don't have these committed to memory, the codes for ceramic capacitor temperature characteristics. Just to make this fun, there are two sets, one for temperature compensating caps and one for caps that claim no such magnificence. For years I thought an X7R was a sports car.  Imagine my disappointment.

These are just screen shots straight out of the Wikipedia article.

http://en.wikipedia.org/wiki/Ceramic_capacitor

Maybe Im a bit denser than usual today but Ive no idea what those charts mean :o

Its no wonder TC is such a black art

Carl

I don't know what this has to do with orange drops.  You study this stuff when selecting caps for loading in an output pi network, DC blocking on PA anodes and h.v. choke bypass.   It matters when the cap has to handle a lot of current on a low frequency like 1800 kc and you don't want a value drift but you don't want to use a big tx mica.  so you have the temp codes for the dielectric; the other factor though is the environmental heating which is of concern if the caps are near the tubes (DC blocking caps).   you have two classes, class I and Class II dielectric and IIRC, Class II is what you want for RF ceramic caps.   

Hi All,

I generally prefer the tubulars to replace papers. It's a fit thing mostly and the performance differences, even on the higher bands, is insignificant in my experience. I dislike OD's in general; appearance, working with them, fitting them in some cases.

I don't have the data and opinion is formed more from observation.

While we on the issue of caps, with replacing lytic's particularly in PS applications, I always stick with the original values. I can give you examples of me selling transmitters in particular to a Hammy Hambone. I had one replace the new caps in all the supplies in a transmitter and then contacted me saying he wanted his money back because the transmitter wouldn't tune, and he couldn't adjust the bias, and other issues. I had him return it to me so I could look at it, since it worked perfectly to spec when it left here.

Sure enough the cap values were all 3 to 4 times higher in the "better" caps he put in. Turns out he had a ground loop issue in his audio gear, and thought the transmitter was humming. The voltages in the transmitter were all about 20% too high (except for the plate), and the idea of even adding series resistance (a bandaid for a bad approach, IMHO) didn't occur to him. Indeed, as it was the transmitter was all out of whack due to the differences in voltages coming out of the PS. He even removed a series resistor added to a diode that replaced a selenium on the bias supply because it wasn't on his schematic.

So I replaced all the capacitors back to the original values (and put back the series resistor in the bias) and everything worked fine. Now this was the case in this specific transmitter, and I know it is not as critical with others. However, I have never understood why values that work in the first place, particularly in choke input PS's with good regulation, are changed to a higher value.

But I'm off topic. Suffice it to say, I do prefer tubulars and find them to work well for bypass & coupling applications. I do tend to use either Polypropylene or even Teflon in some cases though. For circuits requiring stability, I do tend to be more particular and go with ceramics of the higher precision, better tolerance types (C0G or NPO). Z5U and 2E6 ceramic types I avoid completely, due to microphonics issues, particularly in audio stages especially modulators. I discovered in my B&W 5100B what I thought was feedback from my modulation transformer was actually microphonics in the modulator from a few Z5U type ceramics for coupling that found their way in there at some point. For HF or VHF tuned circuits I generally use any good quality silver mica.

Measuring the value of given capacitor to be used, using a heating and cooling method during testing as well, is also highly recommended.

John

What page are you on? We are not talking about high voltage, DC blocking applications or anything like that. The thread is about ... oh well.

Sorry about that.  An explanation is in order.  Here's a longer quote from the article  ( http://en.wikipedia.org/wiki/Ceramic_capacitor )

"Oh well" is right.  Go back and do some more reading.  Read closely.

Just some random notes echoing what others have said...

1.  In the SP-600 Hammarlund seemed to have moved from Black Beauties to ceramic disks.  A friend has an SP-600 that (he claims) was capped at the factory with ceramic disks in place of all of the Black Beauties used in earlier builds.

2.  I don't know about ceramic disks, but surface-mount ceramic caps are certainly microphonic (i.e. piezoelectric effect).  However, in my experience with SMT caps the amount of energy generated via mechanical coupling has been negligable with respect to overall signal levels.  If you're worried about a cap's possible piezoelectric effects in a receiver, just tap the cap-in-question with a pencil (or something non-conductive).  Tap around it, too, in case the piezoelectric effect is due to something else.  But I'd think that mechanical coupling to a leaded cap (rather than an SMT cap) would be minimal, as the leads themselves ought to attenuate the mechanical energy.

3.  In the SMT world, I never use Z5U caps -- the dielectric layer is very thin, and they can more easily develop resistive leakage via micro-cracks.  Been there, bit by that.

4.  The Director of Operations at one of the companies I worked at insisted that we use electrolytics rated to 100 degrees C in lieu of the cheaper 85 degree C equivalent parts, because, he claimed they're more reliable.  Don't know if this is true, or not, but to this day I prefer to use the higher-temp-rated parts when I can get them, and it certainly isn't a bad idea in tube gear, where the temperatures can be higher than temps in solid-state equipment.

5.  For replacing RF bypass caps in a tube receiver I'd naturally opt for ceramic disks in lieu of larger "rolled" caps (with their larger inductance and larger surface area (i.e. stray capacitive coupling)).  In audio applications...depends on what capacitance is needed, and at what voltage (and what is in the junkbox).  Also, when bypassing, a larger capacitance value will almost always work just as well (and perhaps better) than the original value (so I use whatever is in my junkbox that's close).  But pay attention to the cap's purpose in the circuit, as sometimes you don't want to change the capacitance value.

6.  Silver-Mica caps can develop resistive leakage -- this seems to be a problem in the CE-100V phase network, where certain mica caps in the circuit can start to leak and throw off the phase-shift properties of the network (it was a problem in mine, and if I recall correctly others have experienced it, too).  Probably due to the high-voltage across the caps-in-question intensifying silver migration, so I'd recommend caution when using micas in applications with large DC voltages across them.

7.  I have no problem replacing drifting carbon comp resistors with more modern film resistors.  But pay attention to package power and voltage ratings versus application power and voltage.  If voltage rating is marginal, use two (or more) resistors of the same value in series.

(Jon -- if you're ever curious about a cap's ESR, resonant frequency, or other parasitic effects, bring it by the house sometime and we'll put it on the network analyzer.)

- Jeff, k6jca

www.k6jca.blogspot.com

Early 50's high value discs were microphonic at times but also had leakage. The only pencil tapping effect I can think of in over 50 years of repairing radios is those damn Centralab and other brand "Couplates" and they are automatically replaced these days.

Carl

Jeff, We have had a lot of issues with SM electrolytics so going to a higher rating is a good idea. We derated the voltage rating.

This post is way out of line from the original thread, but what the heck. One note of caution... Although the VNA is great for measurements of most run of the mill caps, its not an easy or accurate measurement if you want to measure a really GOOD cap. Incidently, I returned to the original 100 WV Sprague OD that I posted at reply #2 and remeaured it on an LCR Bridge. The measurement was at 120 Hz and 1 kHz. The DF gave a reading of .003 at 1 kHz implying a Q for the cap of 330 and therefore an Rs of 10 ohms. Not to shabby and in the ball park of my original post. I agree on RF bypassing. I think I would rather use a CD or some other form of suitably HV cap whose self resonant frequency is as high as possible and with as high a Q as possible in the RF deck of any tube radio.

Since the topic of Q has been mentioned, let me digress even further from the original topic and mention that Dick, W1QG, has an interesting method of measuring Q using a VNA, which you can find described on his website: http://home.comcast.net/~w1qg/

- Jeff, k6jca

www.k6jca.blogspot.com

Good read and thanks Jeff!

Alan

Because they look so good in point to point work! All this talk about induction and impedance but the real key to Orange Drops is there appearance. Maybe if you're doing strip line or microwave but on my 160 meter AM transmitter they rule.
RF

Actually it's the orange drops, and waxed paper capacitors they prefer.  If I had a dime for every audiophool musician and "boutique" amp builder, I've gotten in a discussion/argument about how orange drops, and waxed paper capacitors make your amps, and your guitars "sound better", I'd be a millionare!  You know what's the worst?, when you tell them that at audio ferquencies, there really is no difference in their reactance frequency/tone-wise, and they respond to you that "there really is a difference - you just have to have a trained ear to hear it."    ::)

Very interesting read from Dick on the VNA technique for measuring Q.

What I miss most about retiring was access to MATLAB and Microwave Office plus several other specialized engineering programs not available to mere mortals at any price.

Or even just an affordable version of Orcad!

I had a copy of Orcad until that PC blew up and I still dont understand why the backup copy on another PC in the home network wouldnt work.

The exact same thing happened to me about two months ago!  My PC with Orcad died a sudden death, and I didn't have a backup copy.  I've since downloaded their free copy, but it is a pitiful replacement, allowing only a small number of components and nets to a design.

Looks like I'll be parsing out new designs onto A-size sheets for the foreseeable future!

- Jeff

I managed to grab a light table that was no longer needed at a past employer. It still comes in handy but the only place I can fit it now is in an unheated attic.

Jeff, Go to LTC and download LT SpiceIV. It is great for simulations and drawing schematics. Similar to orcad.
I can't help you with matlab.

Frank,

    Would you consider chip caps suitable for screen R.F. bypass service? Should they be de-rated?

TIA.
Dave.

Hope for American made caps fades. 

ASC caps apparently are made in the US but are very pricey at least at Allied.  For example, a .01 400V polypropylene costs $3.01.  How many do you want to buy at three bucks a pop?

And I received this response from EFC/Wesco

Hi Jon,
 
Thank you for the inquiry.  I don’t believe any of our distributors do internet retail business.  Are business focus is OEM and we do not typically sell retail to individuals.  Our distributors do not stock our product as it is all made to order.
 
Regards,
Bob

 ;)

Hi Folks,

So here's me.. I've referbed a few radios, a ZTO being the last.  I used the yellow metalized Poly caps.

In looking at the comparisons from one of the suppliers I don't see a whole lot of diff between the yellow and the orange.

Yellow come in:
Metalized Polypropylene Film and Metalized Polyester Film.
The later have self healing properties.

The Yellow polyprop:
# Rated Voltage 630 Volts & 1000 Volts
# Wrapped & Filled, Tubular. Axial
# Available in "hard to find" tube radio mfd/uF sizes such as 0.0005 mfd, 0.002 mfd, 0.005 mfd, etc.
# Excellent for use in tube radios, amps and hi-fi audio equipment where low dissipation / high insulation resistance and excellent long-term stability are desirable.
# Capacitance tolerance of +/- 10%.
# Superior replacement for paper/wax capacitors in tube electronics.
# Made by ISO 9002 & 9001 certified manufacturer / OEM supplier with major customers: CDE, Sony, National, Canon, APC, LG, LEO, Bosch, B&D, Askey and HP.

The Yellow polyester:
# Rated Voltage 630 Volts & 1000 Volts
# Wrapped and Filled, Tubular Axial
# Self-healing property
# Available in "hard to find" pre WWII sizes such as 0.02 MFD, 0.03 MFD, 0.04 MFD, 0.05 MFD, etc.
# Excellent for use in tube radios, amps, hi-fi, etc, where a premium quality capacitor with high stability and reliability are required. Also ideal for telecommunications, industrial and general electronics equipment.
# Made by ISO 9002 & 9001 certified manufacturer / OEM supplier with major customers: CDE, Sony, National, Canon, APC, LG, LEO, Bosch, B&D, Askey and HP.

And lastly the Orange drops which come in two exciting flavors:
Metal-Foil Polypropylene Film and Metalized Polyester Film.  The later has the "self healing trait too.

The orange polyprop:
# Rated Voltage 630 and 1600 Volts
# Orange epoxy coated for superior heat, solvent and moisture resistance.
# Extra long radial leads for easy under chassis installation.
# Superb replacement for old paper/wax capacitors in tube radios, hi-fi and amplifiers, where high stability and high insulation resistance are desirable.
# Excellent stability for High Current / High Frequency / Temp. Compensation Applications.
# Made by ISO 9002 & 9001 certified manufacturer / OEM supplier with major customers: CDE, Sony, National, Canon, APC, LG, LEO, Bosch, B&D, Askey and HP.

The orange polyester:
# Rated Voltage 630 Volts
# Dipped in Orange epoxy for superior heat, humidity and solvent resistance.
# Highly reliable with "self-healing" performance
# Extra long radial leads for easy under chassis installation.
# Excellent for DC blocking, coupling, filtering and by-pass in tube radios, hi-fi, amps and communications equipment.
# Made by ISO 9002 & 9001 certified manufacturer / OEM supplier with major customers: CDE, Sony, National, Canon, APC, LG, LEO, Bosch, B&D, Askey and HP

This info is from the "JUST Radios" website.  ;)

So I have to decide what to order.  (for my NC-183D)  Not that great of difference in the cash output.  humm... I like the way the Yellows go in..  The leads are Axial.  How long does one last compared to the other?  humm...

Reading this tread.. I lean this way for a bit the flop over to the other side.  I just don't know.
Help me Obewan..  ::) Me confused... ;D

Tnx, Scotty

I have NOS mylar caps from the 60's that are still perfect.

Since the OD's and the no name yellow axials are from China you take your chances and I NEVER buy anything from garage operations. Just Mouser and industrial surplus dealers. I dont want any of my gear crapping out or customer returns just cuz I was too cheap to stick with known quality.

Carl

YOU DA MAN! ! ! ! ! ! ! ! ! ! ! ! !

I feel pretty much the same way!

 :)

Good Morning!

Well that would be nice.  But I think many of the manufactures like Cornel Dubilier are a large conglomerate.  Here's an excerpt from there web page.. Don't get me wrong, they have great stuff! 

Quote id from this page:  http://www.cde.com/about/ (http://www.cde.com/about/)

and I quote  "A global company, Cornell Dubilier has ISO-9001 certified manufacturing and distribution facilities in Liberty, SC; New Bedford, MA; Mexicali, Mexico; and Shenzhen, China to serve customers worldwide. Cornell Dubilier products are recognized by DESC, as well as safety agencies UL and CSA. "  End Quote.

The companies I found on the Mouser web page for Capacitors are :
AVX
Cornel Dubilier
Dielectric Labs
Icoxris
Elpac
Kemet.. this on is an open DNS now??
Maxwel Tech.
Murata
Nichicon
Novacap
NTE
Susumu
TDK
Vishay
Wima
Xicon

Look at any of these manufactures and I think you'll find they had to build out of the States.. Just to do what they do with out breaking Environmental Laws.  They can pollute in China and others.. but not here in the States...(well not too much GRIN) :)

So say at Digi Key or Mouser for instance... which brand name do you go for?

I like buying from that little ma and pop site.  They have fresh stock.. and do a great job!  IMO Plus they give you free shrink wrap with every order.  Nice!  Oh. their fast too. I don't work for them by the way.. nor have any interest or ties with them in anyway.  :)

I'm making a list of the caps I need.. and checking it twice!  ::)

Scotty

There is a big difference from buying say Nichicon in China and some no name crap with no paper trail.
The major players have their own factories in China with the same standards as the US, Japan, Taiwan, etc.

For HV 'lytics as used in amps I use only CDE 381LX series made in South Carolina, never had a bad one or any reported failures and Ive been using them since they were first stocked at Mouser maybe 25 years ago.

BTW, you left United Chemi-Con off that Mouser list.

Carl

Dave,
Chip ceramic caps would make a fine bypass as long as the voltage rating is good.
They also make RF chip caps with foil leads. a CKO6 has a chip cap inside.

TNX.

Dave,

For ceramics we always used a voltage de-rating factor of .6 to get the MTBF that we were looking for.  Ceramics caps are very long life caps when used properly.

Joe, W3GMS

I Agree Joe.
You really need to belt a ceramic chip cap to take it out.
CK06s and CR05s also good. I'm running 100 volt CK06s on my 300 watt Erbtec
final at 50 volts supply. Push pull the voltages are well over 100. I'll be swapping to 200V chips soon.
CKR22s are the only ceramic caps I have seen any real failures out of the box.
Even disc ceramics don't usually fail.

I now use Mallory 150s in all my guitar amp kits rather than orange drops. Mouser has them in stock and they are smaller than the Orange Drops and seem every bit as good. I like the axial leads. They used to be yellow but all of the recent ones are now white. Cornell Dublier makes some very high quality tubular caps for high current applications but they are not cheap. For my Class E PWM filter , though, I used them exclusively. You should use what the application calls for!

Pat

 ;)
Looking at some of these caps I see for instance a yellow .5mfd says ETR MET and the value of 504K630V.  The "ETR" must be the brand.  The "MET" is for Metalized Polyester Film i guess.
ETR is ETR electronics, LTD

Subsidiaries:
   

1. Sinhua Electronics (Shanghai) Co., Ltd.

2. ETR Electronics (Singapore) Co., Ltd.

3. ETR Electronics (India) Co., Ltd.

4. ETR Electronics (Hong Kong) Co., Ltd.
Product Items:    

1. Polyester Film Capacitor
2. Polypropylene Film Capacitor
3. Metalized Polyester Film Capacitor
4. Metalized Polypropylene Film Capacitor
5. LEDs & LED Displays
6. Tantalum Capacitors


A Electrolytic has a brand name on it of MIEC  and states the value in MFD.
Here's what I found:
Ling In Electronics Co., Ltd. - Aluminum Electrolytic Capacitor, Electrolytic Capacitor, Middle and High Voltage Capacitor, Car Capacitor.

So.. what am I gonna use?  U bet..  You guessed it!   ;D

I wouldn't hang my hat on the Orange Drops being USA parts - Maybe so but likely not. Sprague no longer exists - Vishay bought them years ago. As someone mentioned earlier - it's not where they are made - its to what the manufacturing  specs are. Lots of USA companies have quality parts made outside the USA.

Cornell Dublier has a plant in Myrtle Beach and Kemet has a plant in Greenwood, SC but most of their stuff is manufactured overseas. I have been inside the Kemet plant. They package the chips into epoxy and add leads but the chips come in in plastic barrels from Mexico and other places.

I used to use the Atom electrolytic caps - they cost three times as much as the Nichicons and United Chemicons and then the price trippled - all for an inferior capacitor that was twice as large and with worse leakage.

Pat

Remember the guy selling Spprague caps at Deerfield? I bet those bags were the last of the USA made caps.

Pat,

I just don't see the Mallory brand at Mouser.

Yup.. with postal rates going up this hobby or profession is going to have a little higher overhead.  Golf anyone? :o

They are on Mouser catalog page 958  - they are actually manufactured by CD now - I am not sure when they started but they are still called Mallory 150s. Don't know if they are manufactured in Myrtle Beach or offshore now.

Pat

I bought several at Spring and Fall. His wife worked there and most were oddball but useable values to anyone that has clue. I also keep his amps running 8)

<raises hand>

Oh, oh, oh!

Pick me! Pick me!!

I know, I know I know!

It's because Lemon Drops won't work!!  :P :P

                 _-_-bear


(ok, sorry, just had to say it...)

Just buy tear drops get over with! HAHAHA!