|
Roy,
We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Tam/WB2TT |
Tam/WB2TT wrote:
Roy, We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Tam/WB2TT Perhaps you were talking about bypass type ceramics, but the posting I responded to: Yes, ceramics are *hopeless* for tuned circuits; I wouldn't trust the black tipped ones, either. You can't beat silver mica but they're a bit hard to find and expensive. made no such qualification, and even specifically mentioned "black tipped" (usually NPO) capacitors. As far as I know, silver micas are getting pretty rare, except maybe for very high power, high current RF applications, if they're being used for even that any more. I doubt if they exist as surface mount parts. The reason silver micas are hard to find and expensive is that they've been made obsolete for nearly all applications by generally superior ceramic types. Roy Lewallen, W7EL |
Tam/WB2TT wrote:
Roy, We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Tam/WB2TT Perhaps you were talking about bypass type ceramics, but the posting I responded to: Yes, ceramics are *hopeless* for tuned circuits; I wouldn't trust the black tipped ones, either. You can't beat silver mica but they're a bit hard to find and expensive. made no such qualification, and even specifically mentioned "black tipped" (usually NPO) capacitors. As far as I know, silver micas are getting pretty rare, except maybe for very high power, high current RF applications, if they're being used for even that any more. I doubt if they exist as surface mount parts. The reason silver micas are hard to find and expensive is that they've been made obsolete for nearly all applications by generally superior ceramic types. Roy Lewallen, W7EL |
In article ,
Tam/WB2TT wrote: Roy, We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Cornell Dubilier / Waldom makes surace mount siler mico caps. You can get them from Digikey for under $10 US. You can get standard PPS film capacitors from several makers. If you don't mind the fact that they are very touchy and have a higher failure rate they may be an option. If you want a little better, you can buy the more costly coated ones. I wouldn't say that you are forced into using ceramic. It is an option you may perfer. -- -- forging knowledge |
In article ,
Tam/WB2TT wrote: Roy, We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Cornell Dubilier / Waldom makes surace mount siler mico caps. You can get them from Digikey for under $10 US. You can get standard PPS film capacitors from several makers. If you don't mind the fact that they are very touchy and have a higher failure rate they may be an option. If you want a little better, you can buy the more costly coated ones. I wouldn't say that you are forced into using ceramic. It is an option you may perfer. -- -- forging knowledge |
On Tue, 23 Mar 2004 14:44:30 +0000 (UTC),
(Ken Smith) wrote: In article , Tam/WB2TT wrote: Roy, We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Cornell Dubilier / Waldom makes surace mount siler mico caps. You can get them from Digikey for under $10 US. The CDE "MC" series of cazapitors are "mica", not "silver mica". The difference is that silver mica caps have to be sealed (dipped) or the silver plating reacts with everything. I'm not sure what plating is used for the "MC" series of surface mount mica. My guess(tm) is aluminum. http://www.cornell-dubilier.com/mica/mica.htm http://www.cornell-dubilier.com/film/hmc.htm http://www.cornell-dubilier.com/catalogs/MC.pdf The big advantages of silver mica is stability, wide temp range, very low dissipation, and tolerance to over voltage spikes. Many years ago, I wasted a month working over an HF xmitter, trying to design out the expensive silver mica and porcelain cazapitors and replace them with cheaper ceramics. It was possible for the low power drivers but a waste of time in areas that had high RF currents or required good stability. A similar cost reduction exercise was also being done on the automagic antenna tuner (by someone else) with similar results. The project ended when someone suggested using high temp silver solder to prevent the ceramic caps from reflowing their solder connections and falling off the board. I guess(tm) the reason that silver mica caps are difficult to find is that there are few companies producing high power RF products as compared to the huge number of low power RF products. It's not a big market that probably can only support a few specialty component vendors. -- Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060 (831)421-6491 pgr (831)336-2558 home http://www.LearnByDestroying.com AE6KS |
On Tue, 23 Mar 2004 14:44:30 +0000 (UTC),
(Ken Smith) wrote: In article , Tam/WB2TT wrote: Roy, We were talking about bypass type ceramics. See the 3/20 10:13AM posting. BTW, I think with SM you are pretty much forced into using ceramics. Cornell Dubilier / Waldom makes surace mount siler mico caps. You can get them from Digikey for under $10 US. The CDE "MC" series of cazapitors are "mica", not "silver mica". The difference is that silver mica caps have to be sealed (dipped) or the silver plating reacts with everything. I'm not sure what plating is used for the "MC" series of surface mount mica. My guess(tm) is aluminum. http://www.cornell-dubilier.com/mica/mica.htm http://www.cornell-dubilier.com/film/hmc.htm http://www.cornell-dubilier.com/catalogs/MC.pdf The big advantages of silver mica is stability, wide temp range, very low dissipation, and tolerance to over voltage spikes. Many years ago, I wasted a month working over an HF xmitter, trying to design out the expensive silver mica and porcelain cazapitors and replace them with cheaper ceramics. It was possible for the low power drivers but a waste of time in areas that had high RF currents or required good stability. A similar cost reduction exercise was also being done on the automagic antenna tuner (by someone else) with similar results. The project ended when someone suggested using high temp silver solder to prevent the ceramic caps from reflowing their solder connections and falling off the board. I guess(tm) the reason that silver mica caps are difficult to find is that there are few companies producing high power RF products as compared to the huge number of low power RF products. It's not a big market that probably can only support a few specialty component vendors. -- Jeff Liebermann 150 Felker St #D Santa Cruz CA 95060 (831)421-6491 pgr (831)336-2558 home http://www.LearnByDestroying.com AE6KS |
Roy Lewallen wrote in message ...
Paul Burridge wrote: Yes, ceramics are *hopeless* for tuned circuits; I wouldn't trust the black tipped ones, either. You can't beat silver mica but they're a bit hard to find and expensive. I strongly disagree with this. I've successfully used ceramic capacitors many times for both high and low Q tuned circuits from HF to UHF. Years ago, I found that NPO ceramics were decidedly superior to silver micas for temperature stability, so I use them exclusively for VFO tank circuits. .... In agreement and support of what Roy wrote, I'd toss out some additional notes: o I've seen (been the vicitm of?) silvered micas that exhibit random tiny jumps in capacitance, which is a really bad thing in oscillators. o You can get X7R dielectric SMT caps in SMT up to at least 0.1uF in 0603 size, for low voltage ratings. See manufacturers' data sheets for the largest currently available values. I suspect reliability suffers if you try to use ones with too high a C*V/unit volume, though. o As someone recently pointed out to me (Sphero, I think it was), you can get C0G up to 0.1uF in SMT--I think he said in 1206 size. o It's possible to get 1% C0G caps. o A C0G is probably about the cheapest electronic component you'll find with a maximum 30ppm/C temperature coefficient (though beware, some C0Gs I've seen are rated up to 60ppm/C). o There are special high-Q ceramics that are better than standard C0Gs for use at microwave frequencies...generally above 1GHz. Cheers, Tom |
Roy Lewallen wrote in message ...
Paul Burridge wrote: Yes, ceramics are *hopeless* for tuned circuits; I wouldn't trust the black tipped ones, either. You can't beat silver mica but they're a bit hard to find and expensive. I strongly disagree with this. I've successfully used ceramic capacitors many times for both high and low Q tuned circuits from HF to UHF. Years ago, I found that NPO ceramics were decidedly superior to silver micas for temperature stability, so I use them exclusively for VFO tank circuits. .... In agreement and support of what Roy wrote, I'd toss out some additional notes: o I've seen (been the vicitm of?) silvered micas that exhibit random tiny jumps in capacitance, which is a really bad thing in oscillators. o You can get X7R dielectric SMT caps in SMT up to at least 0.1uF in 0603 size, for low voltage ratings. See manufacturers' data sheets for the largest currently available values. I suspect reliability suffers if you try to use ones with too high a C*V/unit volume, though. o As someone recently pointed out to me (Sphero, I think it was), you can get C0G up to 0.1uF in SMT--I think he said in 1206 size. o It's possible to get 1% C0G caps. o A C0G is probably about the cheapest electronic component you'll find with a maximum 30ppm/C temperature coefficient (though beware, some C0Gs I've seen are rated up to 60ppm/C). o There are special high-Q ceramics that are better than standard C0Gs for use at microwave frequencies...generally above 1GHz. Cheers, Tom |
|
|
Thanks, Jeff and Ken, for the info about modern mica capacitors. I
haven't dealt with the high-power RF world since a previous life as a radar technician, long ago now, so hadn't realized they're still preferred for some applications. Roy Lewallen, W7EL |
Thanks, Jeff and Ken, for the info about modern mica capacitors. I
haven't dealt with the high-power RF world since a previous life as a radar technician, long ago now, so hadn't realized they're still preferred for some applications. Roy Lewallen, W7EL |
Sorry to say that I don't have a recommendation for a vendor for small
quantities of the 1% parts, but how many did you need? I know I have some of the 1.0nF 0805s, and might get lucky on the others too. I have a whole bunch of some value fairly close to 4.7nF, and I'm just not remembering which it is. Might be 3.9, might be 5.6 or 6.8, if not 4.7. Anyway, if you're not looking for too many, I'd be happy to help you out. But even if you buy a bunch of 5% ones, that might not be a bad way to go. I just measured about 20 470pF 5% ones I happened to have on my desk, and they were between 468 and 488, about +/-2% from the midpoint. I'm not remembering where I saw the 60ppm number. It surprised me, though, and I made a mental note to always check the specs on the caps I'm using, if it's important. It was fairly recently, and it was a surprise because I had always equated 30ppm with NPO/C0G. I just did a Google search for "C0G 60ppm" and got several hits, but in looking at some of them, I found 60ppm associated only with NPO, and not with C0G, so perhaps I'm wrong about C0Gs, at least. I'd be happy to be wrong in this case! A little further looking suggests that EIA CG or C0G officially must be +/-30ppm max, and EIA CH is +/-60ppm. Cheers, Tom "Fred Bartoli" r_AndThisToo wrote in message ... "Tom Bruhns" a écrit dans le message news: ... snip o It's possible to get 1% C0G caps. Tom, do you know any supplier of such for proto quantity ? I need 1n/0805 and 4.7n/1206. I can't find any of these and I feel I'll have to buy a handfull of 5%, sort them and maybe adapt the resitors values to what the lot would kindly give me. Not very entertaining. o A C0G is probably about the cheapest electronic component you'll find with a maximum 30ppm/C temperature coefficient (though beware, some C0Gs I've seen are rated up to 60ppm/C). Do you have any name ? Thanks, Fred. |
Sorry to say that I don't have a recommendation for a vendor for small
quantities of the 1% parts, but how many did you need? I know I have some of the 1.0nF 0805s, and might get lucky on the others too. I have a whole bunch of some value fairly close to 4.7nF, and I'm just not remembering which it is. Might be 3.9, might be 5.6 or 6.8, if not 4.7. Anyway, if you're not looking for too many, I'd be happy to help you out. But even if you buy a bunch of 5% ones, that might not be a bad way to go. I just measured about 20 470pF 5% ones I happened to have on my desk, and they were between 468 and 488, about +/-2% from the midpoint. I'm not remembering where I saw the 60ppm number. It surprised me, though, and I made a mental note to always check the specs on the caps I'm using, if it's important. It was fairly recently, and it was a surprise because I had always equated 30ppm with NPO/C0G. I just did a Google search for "C0G 60ppm" and got several hits, but in looking at some of them, I found 60ppm associated only with NPO, and not with C0G, so perhaps I'm wrong about C0Gs, at least. I'd be happy to be wrong in this case! A little further looking suggests that EIA CG or C0G officially must be +/-30ppm max, and EIA CH is +/-60ppm. Cheers, Tom "Fred Bartoli" r_AndThisToo wrote in message ... "Tom Bruhns" a écrit dans le message news: ... snip o It's possible to get 1% C0G caps. Tom, do you know any supplier of such for proto quantity ? I need 1n/0805 and 4.7n/1206. I can't find any of these and I feel I'll have to buy a handfull of 5%, sort them and maybe adapt the resitors values to what the lot would kindly give me. Not very entertaining. o A C0G is probably about the cheapest electronic component you'll find with a maximum 30ppm/C temperature coefficient (though beware, some C0Gs I've seen are rated up to 60ppm/C). Do you have any name ? Thanks, Fred. |
According to information I have, the C0 means zero nominal tempco, and
the G means +/-30 ppm. There are various other letters (unfortunately not in order) representing tolerances from +/-10 to +/-2500 ppm. (There's also the letter O meaning "not specified" and P meaning "see applicable specification".) +/-60 is H, so a 0 +/-60 ppm part would be C0H, not C0G. This series of suffix letters is used for temperature compensating types (e.g., M7 (P100), R2 (N220)) as well, so R2G would be -220 +/-30 ppm, M7H would be +100 +/-60 pmm, etc. Roy Lewallen, W7EL Tom Bruhns wrote: Sorry to say that I don't have a recommendation for a vendor for small quantities of the 1% parts, but how many did you need? I know I have some of the 1.0nF 0805s, and might get lucky on the others too. I have a whole bunch of some value fairly close to 4.7nF, and I'm just not remembering which it is. Might be 3.9, might be 5.6 or 6.8, if not 4.7. Anyway, if you're not looking for too many, I'd be happy to help you out. But even if you buy a bunch of 5% ones, that might not be a bad way to go. I just measured about 20 470pF 5% ones I happened to have on my desk, and they were between 468 and 488, about +/-2% from the midpoint. I'm not remembering where I saw the 60ppm number. It surprised me, though, and I made a mental note to always check the specs on the caps I'm using, if it's important. It was fairly recently, and it was a surprise because I had always equated 30ppm with NPO/C0G. I just did a Google search for "C0G 60ppm" and got several hits, but in looking at some of them, I found 60ppm associated only with NPO, and not with C0G, so perhaps I'm wrong about C0Gs, at least. I'd be happy to be wrong in this case! A little further looking suggests that EIA CG or C0G officially must be +/-30ppm max, and EIA CH is +/-60ppm. Cheers, Tom |
According to information I have, the C0 means zero nominal tempco, and
the G means +/-30 ppm. There are various other letters (unfortunately not in order) representing tolerances from +/-10 to +/-2500 ppm. (There's also the letter O meaning "not specified" and P meaning "see applicable specification".) +/-60 is H, so a 0 +/-60 ppm part would be C0H, not C0G. This series of suffix letters is used for temperature compensating types (e.g., M7 (P100), R2 (N220)) as well, so R2G would be -220 +/-30 ppm, M7H would be +100 +/-60 pmm, etc. Roy Lewallen, W7EL Tom Bruhns wrote: Sorry to say that I don't have a recommendation for a vendor for small quantities of the 1% parts, but how many did you need? I know I have some of the 1.0nF 0805s, and might get lucky on the others too. I have a whole bunch of some value fairly close to 4.7nF, and I'm just not remembering which it is. Might be 3.9, might be 5.6 or 6.8, if not 4.7. Anyway, if you're not looking for too many, I'd be happy to help you out. But even if you buy a bunch of 5% ones, that might not be a bad way to go. I just measured about 20 470pF 5% ones I happened to have on my desk, and they were between 468 and 488, about +/-2% from the midpoint. I'm not remembering where I saw the 60ppm number. It surprised me, though, and I made a mental note to always check the specs on the caps I'm using, if it's important. It was fairly recently, and it was a surprise because I had always equated 30ppm with NPO/C0G. I just did a Google search for "C0G 60ppm" and got several hits, but in looking at some of them, I found 60ppm associated only with NPO, and not with C0G, so perhaps I'm wrong about C0Gs, at least. I'd be happy to be wrong in this case! A little further looking suggests that EIA CG or C0G officially must be +/-30ppm max, and EIA CH is +/-60ppm. Cheers, Tom |
I read in sci.electronics.design that Roy Lewallen
wrote (in ) about 'Extracting the 5th Harmonic', on Tue, 23 Mar 2004: This series of suffix letters is used for temperature compensating types (e.g., M7 (P100), R2 (N220)) as well, so R2G would be -220 +/-30 ppm, M7H would be +100 +/-60 pmm, etc. What does D0G mean, if anything? -- Regards, John Woodgate, OOO - Own Opinions Only. The good news is that nothing is compulsory. The bad news is that everything is prohibited. http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk |
I read in sci.electronics.design that Roy Lewallen
wrote (in ) about 'Extracting the 5th Harmonic', on Tue, 23 Mar 2004: This series of suffix letters is used for temperature compensating types (e.g., M7 (P100), R2 (N220)) as well, so R2G would be -220 +/-30 ppm, M7H would be +100 +/-60 pmm, etc. What does D0G mean, if anything? -- Regards, John Woodgate, OOO - Own Opinions Only. The good news is that nothing is compulsory. The bad news is that everything is prohibited. http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk |
John Woodgate wrote:
I read in sci.electronics.design that Roy Lewallen wrote (in ) about 'Extracting the 5th Harmonic', on Tue, 23 Mar 2004: This series of suffix letters is used for temperature compensating types (e.g., M7 (P100), R2 (N220)) as well, so R2G would be -220 +/-30 ppm, M7H would be +100 +/-60 pmm, etc. What does D0G mean, if anything? I don't find D0 listed in my reference. The only one I have with a 0 number is C0, nominally zero tempco. So if it is a legitimate capacitor designation, I don't know what it is. Getting kind of close to April 1, so maybe it's short for D0G B0NE, a kind of ceramic capcitor? Roy Lewallen, W7EL |
John Woodgate wrote:
I read in sci.electronics.design that Roy Lewallen wrote (in ) about 'Extracting the 5th Harmonic', on Tue, 23 Mar 2004: This series of suffix letters is used for temperature compensating types (e.g., M7 (P100), R2 (N220)) as well, so R2G would be -220 +/-30 ppm, M7H would be +100 +/-60 pmm, etc. What does D0G mean, if anything? I don't find D0 listed in my reference. The only one I have with a 0 number is C0, nominally zero tempco. So if it is a legitimate capacitor designation, I don't know what it is. Getting kind of close to April 1, so maybe it's short for D0G B0NE, a kind of ceramic capcitor? Roy Lewallen, W7EL |
"Fred Bartoli" r_AndThisToo wrote in message ...
.... Tom, do you know any supplier of such for proto quantity ? I need 1n/0805 and 4.7n/1206. I can't find any of these and I feel I'll have to buy a handfull of 5%, sort them and maybe adapt the resitors values to what the lot would kindly give me. Not very entertaining. Followup on my previous offer: I have quantities of 3.9nF, 6.8nF and 8.2nF 1%, and a decent supply of 1.0nF 1%, but no 4.7nF rated at 1%. If I had the task of finding a dozen or two 1% 4.7nF parts, I'd just get a hundred 5% parts, make up a set of maybe ten bins (or just divide a sheet of paper into ten or twenty areas) in 1% increments, and sit down for about five minutes with my capacitance tweezers. The sorting goes really fast that way: pick up a part with the tweezers, read its value and drop it into the right bin. Cheers, Tom |
"Fred Bartoli" r_AndThisToo wrote in message ...
.... Tom, do you know any supplier of such for proto quantity ? I need 1n/0805 and 4.7n/1206. I can't find any of these and I feel I'll have to buy a handfull of 5%, sort them and maybe adapt the resitors values to what the lot would kindly give me. Not very entertaining. Followup on my previous offer: I have quantities of 3.9nF, 6.8nF and 8.2nF 1%, and a decent supply of 1.0nF 1%, but no 4.7nF rated at 1%. If I had the task of finding a dozen or two 1% 4.7nF parts, I'd just get a hundred 5% parts, make up a set of maybe ten bins (or just divide a sheet of paper into ten or twenty areas) in 1% increments, and sit down for about five minutes with my capacitance tweezers. The sorting goes really fast that way: pick up a part with the tweezers, read its value and drop it into the right bin. Cheers, Tom |
On Wed, 24 Mar 2004 07:08:11 +0000, John Woodgate
wrote: What does D0G mean, if anything? Jordan. :- -- The BBC: Licensed at public expense to spread lies. |
On Wed, 24 Mar 2004 07:08:11 +0000, John Woodgate
wrote: What does D0G mean, if anything? Jordan. :- -- The BBC: Licensed at public expense to spread lies. |
Yet another followup...
Mouser lists a few 1% values. You could give them a call to see if they happen to have the ones you want. But the prices listed for the 1% values are so high I'd still consider selecting from 5% values, for hobby work. (I'm reading this in r.r.a.homebrew, but I see it's also crossposted to sci.electronics.design...so maybe paying a few dollars each isn't prohibitive.) Cheers, Tom (Tom Bruhns) wrote in message om... "Fred Bartoli" r_AndThisToo wrote in message ... ... Tom, do you know any supplier of such for proto quantity ? I need 1n/0805 and 4.7n/1206. I can't find any of these and I feel I'll have to buy a handfull of 5%, sort them and maybe adapt the resitors values to what the lot would kindly give me. Not very entertaining. Followup on my previous offer: I have quantities of 3.9nF, 6.8nF and 8.2nF 1%, and a decent supply of 1.0nF 1%, but no 4.7nF rated at 1%. If I had the task of finding a dozen or two 1% 4.7nF parts, I'd just get a hundred 5% parts, make up a set of maybe ten bins (or just divide a sheet of paper into ten or twenty areas) in 1% increments, and sit down for about five minutes with my capacitance tweezers. The sorting goes really fast that way: pick up a part with the tweezers, read its value and drop it into the right bin. Cheers, Tom |
|
|
|
|
Go to Agilent home page, search for 16334A.
I also have a homebrew design that works fine, resolution to as little as 0.01pF, but haven't had a chance to move it from the breadboard stage to a more finished implementation. It will be tweezers that connect to a readout box through a single coaxial cable (RG-174-type). There can be interchangeable "heads" -- tweezers, spring clips, etc. If/when I get enough round tuits to finish it up, I'll post/publish something on it. Cheers, Tom Rex wrote in message . .. On 23 Mar 2004 23:50:18 -0800, (Tom Bruhns) wrote: with my capacitance tweezers That sounds interesting. I have used regular tweezers to slip sm caps into a fixture I made. This goes pretty fast except when I accidentally twang a part into some far and unknown location. Did you make these youself? Can you give a quick description. I can think how I could make low capacitance tweezers, but not how to flexibly connect them to a capacitance meter and get consistant readings. |
Go to Agilent home page, search for 16334A.
I also have a homebrew design that works fine, resolution to as little as 0.01pF, but haven't had a chance to move it from the breadboard stage to a more finished implementation. It will be tweezers that connect to a readout box through a single coaxial cable (RG-174-type). There can be interchangeable "heads" -- tweezers, spring clips, etc. If/when I get enough round tuits to finish it up, I'll post/publish something on it. Cheers, Tom Rex wrote in message . .. On 23 Mar 2004 23:50:18 -0800, (Tom Bruhns) wrote: with my capacitance tweezers That sounds interesting. I have used regular tweezers to slip sm caps into a fixture I made. This goes pretty fast except when I accidentally twang a part into some far and unknown location. Did you make these youself? Can you give a quick description. I can think how I could make low capacitance tweezers, but not how to flexibly connect them to a capacitance meter and get consistant readings. |
In
That sounds interesting. I have used regular tweezers to slip sm caps into a fixture I made. This goes pretty fast except when I accidentally twang a part into some far and unknown location. Did you make these youself? Can you give a quick description. I can think how I could make low capacitance tweezers, but not how to flexibly connect them to a capacitance meter and get consistant readings. Wayne Kerr used to make a VHF admittance bridge which serves well for measuring chips, and they can be bought for less than £100 (813? bridge) in UK leaded low value ceramic and silvered mica have their Tc and loss modified by the effect of the encapsulation material. It therefore follows that chips npo chip will be better. -- ddwyer |
In
That sounds interesting. I have used regular tweezers to slip sm caps into a fixture I made. This goes pretty fast except when I accidentally twang a part into some far and unknown location. Did you make these youself? Can you give a quick description. I can think how I could make low capacitance tweezers, but not how to flexibly connect them to a capacitance meter and get consistant readings. Wayne Kerr used to make a VHF admittance bridge which serves well for measuring chips, and they can be bought for less than £100 (813? bridge) in UK leaded low value ceramic and silvered mica have their Tc and loss modified by the effect of the encapsulation material. It therefore follows that chips npo chip will be better. -- ddwyer |
I read in sci.electronics.design that ddwyer
wrote (in ) about 'Extracting the 5th Harmonic', on Sat, 27 Mar 2004: Wayne Kerr used to make a VHF admittance bridge which serves well for measuring chips, and they can be bought for less than £100 (813? bridge) You might well be able to extend its frequency range downwards, too. IIRC, it needs some very small 1 uF capacitors, which simply weren't available when it was manufactured. There is an associated transistor test set which is a walking disaster. -- Regards, John Woodgate, OOO - Own Opinions Only. The good news is that nothing is compulsory. The bad news is that everything is prohibited. http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk |
I read in sci.electronics.design that ddwyer
wrote (in ) about 'Extracting the 5th Harmonic', on Sat, 27 Mar 2004: Wayne Kerr used to make a VHF admittance bridge which serves well for measuring chips, and they can be bought for less than £100 (813? bridge) You might well be able to extend its frequency range downwards, too. IIRC, it needs some very small 1 uF capacitors, which simply weren't available when it was manufactured. There is an associated transistor test set which is a walking disaster. -- Regards, John Woodgate, OOO - Own Opinions Only. The good news is that nothing is compulsory. The bad news is that everything is prohibited. http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk |
Avery Fineman wrote:
In article , Peter John Lawton writes: The above will hold true at any fundamental frequency provided the rise and fall times are equal and each equal to 0.02 times the repetition period. Those numbers will change given faster or slower rise/fall times. All db calculated as 20 x Log (voltage). Width is determined at the baseline, not the 50% amplitude point. Len Anderson retired (from regular hours) electronic engineer person I wonder what happens to these numbers as the rise/fall time tends to zero? The harmonic content will increase...but also show dips depending on the percentage width relative to the period. I could present those (takes only minutes to run the program and transcribe the results) but that is academic only. The rise and fall times will NOT be zero due to the repetition frequency being high (repetition time short). Consider that a 3 MHz waveform has a period of 333 1/3 nSec and that Paul is using a TTL family inverter to make the square wave. Even with a Schmitt trigger inverter the t_r and t_f are going to be finite, possibly 15 nSec with a fast device (and some capacitive loading or semi-resonant whatever to mess with on- and off-times). 15 nSec is 4.5% of the repetition period, quite finite...more than I showed on the small table given previously. I'm sure someone out there wants to argue minutae on numbers but what is being discussed is a squarish waveform with a repetition frequency in the low HF range. Periods are valued in nanoSeconds and the on/off times of squaring devices are ALSO in nanoSeconds. There's just NOT going to be any sort of "zero" on/off times with practical logic devices used by hobbyists. I just wondered from a theoretical point of view what the program would say about the harmonic content as you decreased the values you put into it for t_r and t_f. What is not intuitive to me (and to others) is that harmonic energy of a rectangular waveform drops drastically by the 5th harmonic and is certainly lower than "obvious" numbers bandied about. This is connected with my question. I am pondering why the energy available for higher harmonics is less than for the fundamental and also how your program works out this energy. But, also mentioned before by others is that shortening the rect- angular waveshape DOES increase the 5th harmonic, as evident by the approximate 12 db increase at 40 to 35 percent of the repetition period. Its like pushing the baby on the swing in the park, you only need to give it the occasional push or pull in the right direction. A 5f resonator gets has to go for 2.5 cycles in between refuelling from a square-wave (1:1) of frequency f. With a mark/space ratio of 2:3 it looks to me as though the 'pull' on the 5f resonator as the rectangular wave drops will take out all the energy that was put in on the preceding 'push' - so no 5th harmonic. On the other hand, 1.5:3.5 (30%) should be just as good for 5th as 1:1 (but no better). All that's assuming a zero t_r and t_f. In practice it must be that the energy transfer to a resonator depends on these times. It seems that the energy transfer is not so great to a faster resonator. Pursuing the swing analogy, your arm has to move faster than the swing if you're going to add to the swings energy. An equivalent shortening happens in vacuum tube multipliers through biasing (self, fixed, or both) and that can be adjustable along with the drive level. It's not quite the same with bipolars since the overdrive effects are more saturation than in the self- bias conditions of tubes. It's close, though. From all indications of the Fourier series results, there's a definite reason why so few multipliers went beyond tripling. The amount of energy (relative to fundamental and taking into account the finite rise and fall times) of 4th and higher harmonics just isn't as much as intuition would have everyone believe! What do you mean by intuition here? My intuition suggests to me that as the rise and fall times get shorter, the energy available for the harmonics approaches that for the fundamental. In other words, as a square wave approaches perfection it gains the ability to stimulate all odd harmonic resonators equally - but surely that can't be right. Possibly the energy transfer to a f and (say) 5f resonator approaches the same value but the 5f resonator loses 5 energy at five times the rate of the f one, assuming equal Q. Peter Len Anderson retired (from regular hours) electronic engineer person |
Avery Fineman wrote:
In article , Peter John Lawton writes: The above will hold true at any fundamental frequency provided the rise and fall times are equal and each equal to 0.02 times the repetition period. Those numbers will change given faster or slower rise/fall times. All db calculated as 20 x Log (voltage). Width is determined at the baseline, not the 50% amplitude point. Len Anderson retired (from regular hours) electronic engineer person I wonder what happens to these numbers as the rise/fall time tends to zero? The harmonic content will increase...but also show dips depending on the percentage width relative to the period. I could present those (takes only minutes to run the program and transcribe the results) but that is academic only. The rise and fall times will NOT be zero due to the repetition frequency being high (repetition time short). Consider that a 3 MHz waveform has a period of 333 1/3 nSec and that Paul is using a TTL family inverter to make the square wave. Even with a Schmitt trigger inverter the t_r and t_f are going to be finite, possibly 15 nSec with a fast device (and some capacitive loading or semi-resonant whatever to mess with on- and off-times). 15 nSec is 4.5% of the repetition period, quite finite...more than I showed on the small table given previously. I'm sure someone out there wants to argue minutae on numbers but what is being discussed is a squarish waveform with a repetition frequency in the low HF range. Periods are valued in nanoSeconds and the on/off times of squaring devices are ALSO in nanoSeconds. There's just NOT going to be any sort of "zero" on/off times with practical logic devices used by hobbyists. I just wondered from a theoretical point of view what the program would say about the harmonic content as you decreased the values you put into it for t_r and t_f. What is not intuitive to me (and to others) is that harmonic energy of a rectangular waveform drops drastically by the 5th harmonic and is certainly lower than "obvious" numbers bandied about. This is connected with my question. I am pondering why the energy available for higher harmonics is less than for the fundamental and also how your program works out this energy. But, also mentioned before by others is that shortening the rect- angular waveshape DOES increase the 5th harmonic, as evident by the approximate 12 db increase at 40 to 35 percent of the repetition period. Its like pushing the baby on the swing in the park, you only need to give it the occasional push or pull in the right direction. A 5f resonator gets has to go for 2.5 cycles in between refuelling from a square-wave (1:1) of frequency f. With a mark/space ratio of 2:3 it looks to me as though the 'pull' on the 5f resonator as the rectangular wave drops will take out all the energy that was put in on the preceding 'push' - so no 5th harmonic. On the other hand, 1.5:3.5 (30%) should be just as good for 5th as 1:1 (but no better). All that's assuming a zero t_r and t_f. In practice it must be that the energy transfer to a resonator depends on these times. It seems that the energy transfer is not so great to a faster resonator. Pursuing the swing analogy, your arm has to move faster than the swing if you're going to add to the swings energy. An equivalent shortening happens in vacuum tube multipliers through biasing (self, fixed, or both) and that can be adjustable along with the drive level. It's not quite the same with bipolars since the overdrive effects are more saturation than in the self- bias conditions of tubes. It's close, though. From all indications of the Fourier series results, there's a definite reason why so few multipliers went beyond tripling. The amount of energy (relative to fundamental and taking into account the finite rise and fall times) of 4th and higher harmonics just isn't as much as intuition would have everyone believe! What do you mean by intuition here? My intuition suggests to me that as the rise and fall times get shorter, the energy available for the harmonics approaches that for the fundamental. In other words, as a square wave approaches perfection it gains the ability to stimulate all odd harmonic resonators equally - but surely that can't be right. Possibly the energy transfer to a f and (say) 5f resonator approaches the same value but the 5f resonator loses 5 energy at five times the rate of the f one, assuming equal Q. Peter Len Anderson retired (from regular hours) electronic engineer person |
In article , Peter John Lawton
writes: Avery Fineman wrote: In article , Peter John Lawton writes: The above will hold true at any fundamental frequency provided the rise and fall times are equal and each equal to 0.02 times the repetition period. Those numbers will change given faster or slower rise/fall times. All db calculated as 20 x Log (voltage). Width is determined at the baseline, not the 50% amplitude point. Len Anderson retired (from regular hours) electronic engineer person I wonder what happens to these numbers as the rise/fall time tends to zero? The harmonic content will increase...but also show dips depending on the percentage width relative to the period. I could present those (takes only minutes to run the program and transcribe the results) but that is academic only. The rise and fall times will NOT be zero due to the repetition frequency being high (repetition time short). Consider that a 3 MHz waveform has a period of 333 1/3 nSec and that Paul is using a TTL family inverter to make the square wave. Even with a Schmitt trigger inverter the t_r and t_f are going to be finite, possibly 15 nSec with a fast device (and some capacitive loading or semi-resonant whatever to mess with on- and off-times). 15 nSec is 4.5% of the repetition period, quite finite...more than I showed on the small table given previously. I'm sure someone out there wants to argue minutae on numbers but what is being discussed is a squarish waveform with a repetition frequency in the low HF range. Periods are valued in nanoSeconds and the on/off times of squaring devices are ALSO in nanoSeconds. There's just NOT going to be any sort of "zero" on/off times with practical logic devices used by hobbyists. I just wondered from a theoretical point of view what the program would say about the harmonic content as you decreased the values you put into it for t_r and t_f. The harmonic values will change, approaching that of an ideal square wave. That's a truism. With zero rise and fall it IS the same as an ideal square wave. There's NO accurate little formula, saying, or myth that will predict any particular harmonic value. That's the reason for using nice, very quick number-crunching computer programs. What is not intuitive to me (and to others) is that harmonic energy of a rectangular waveform drops drastically by the 5th harmonic and is certainly lower than "obvious" numbers bandied about. This is connected with my question. I am pondering why the energy available for higher harmonics is less than for the fundamental and also how your program works out this energy. My program was developed while at RCA Corporation, specifically in the time period of winter 1973-1974 using the core of three ideal waveforms: rectangular, rising triangle, falling triangle. They relate to a singular waveform using a time-delay formula multiplier so that the rising triangle butts up to (in time) to the start of the rectangular waveform and the falling triangle starts at the end of the rectangular waveform. Entry is rise-time (the rising triangle), fall-time (the falling triangle), and 50% amplitude pulse width which is the rectangular waveform length and the length of the rising and falling triangles adjusted for their inputted times. [draw it out to see it better] Each basic waveform generates its own Fourier coefficient set. All sets are simply added algebraically. Mathematically okay to do that. A quick form of proof of that is to use a simple frequency-to-time transform that works at each specified point in time along the repetition period of the waveform. The original was a time-to- frequency transform, mathematically different than the opposite. If a reconstruction of the frequency-to-time results in the original entry specifications, then it is called accurate enough. I didn't derive the reconstruction transform since it was already in a book. Neither did I derive any of the basic ideal waveforms which were already in the ITT Blue Bible. The delay multiplier used to set rise, fall, and 50% width was another book value, simplified to faster calculation simplicity because the original was a math problem thing with more terms than needed. As to WHY of the energy distribution, that's up to any person who has the textbook formulas and math smarts to fool around with. I can't sum that up in one message. I doubt anyone can. I do know this: Using the formulas and the program, then setting up a test with careful adjustments of a pulse generator and using a well-calibrated spectrum analyzer, the numbers agree within the tolerances of the analyzer calibration. To me, and lots of others, that is all the proof needed. Beyond that, its too much time and nobody paying me to do this... Its like pushing the baby on the swing in the park, you only need to give it the occasional push or pull in the right direction. A 5f resonator gets has to go for 2.5 cycles in between refuelling from a square-wave (1:1) of frequency f. Use any analogue you want. I don't agree with the above, but feel free and I not going further on that... What do you mean by intuition here? My intuition suggests to me that as the rise and fall times get shorter, the energy available for the harmonics approaches that for the fundamental. In other words, as a square wave approaches perfection it For any ideal rectangular shape, the harmonic energies have a (SinX / X) locus. That's explained in textbooks also. Harmonics of a repetitive waveform Fourier transform will NEVER have more energy than the fundamental. That's also basic book stuff. If the rise and/or fall times are finite, the harmonics will drop their energy levels compared to the zero rise and fall time ideals. As the rise and fall times get longer and longer the harmonic energy gets less and less. By the time one gets to a sinusoid waveshape, there are NO harmonics in any Fourier transform, its all fundamental frequency (1 / repetition-period). Recess. Len Anderson retired (from regular hours) electronic engineer person |
| All times are GMT +1. The time now is 05:31 AM. |
Powered by vBulletin® Copyright ©2000 - 2025, Jelsoft Enterprises Ltd.
RadioBanter.com