I think Steve has a valid general point- just as there is a difference between professional designers and amateurs like me, there is also a difference in flexibility. But in the latter case, it's advantage to the amateur. Considerations of manufacturability and universal compatibility can take a back seat to pure sonics and esthetics.
Hi SY,
Being the proud animals that we are, we like to compare our designs to others and commercial stuff also. How in the devil can you do that if your gear works on a completely different standard? Beats me. I guess it makes us safe from comparison, right?
Hi Steve,
-Chris
I agree. However, I like to be practical too.
Being the proud animals that we are, we like to compare our designs to others and commercial stuff also. How in the devil can you do that if your gear works on a completely different standard? Beats me. I guess it makes us safe from comparison, right?
Hi Steve,
I have no problem about that statement. I do tend to think globally as it's been the way I've looked at things for 30 years +. When you service equipment, it teaches you (I hope) to think logically.
I agree. But is that any reason to shoot yourself in the foot by making things that don't play nice with other items? That is a design choice and you are free to do things that way. No skin off my nose. I do think it's a silly way to live.
Well, that's because I have dealt with commercial manufactures for most of my life. Is there anything wrong with that? I just understand what some of their concerns and issues are. For myself, I design to North American hydro (mains) standards and signals standards. I also design amps to drive 4 ~ 8 ohm nominal impedances with headroom. Why on earth would you design an entire system on it's own standard. Each piece will probably be used with normal consumer equipment until all the bits are done. So I still don't understand where you are coming from. Sorry, you're a mystery to me. That's okay though, I'm sure you have a standard mirror to look into for company.
-Chris
SY said:
Exactly.
I think DIY is about more than just saving money and/or a sense of accomplishment. It's also about freedom, and not necessarily having to accept that which the market has ultimately forced upon commercial manufacturers. At least those who wish to sell more than a few pieces of gear.
se
Hi Steve,
I simply stated that there are some things we can't control and some that we have some influence over. Do your best when you design.
Or, give up and built everything out of plastic or cardboard. This I did not suggest. It's just the other extreme.
So, the only thing I am wondering is why you singled my comments out. It does not appear that we are in disagreement at all.
-Chris
Edit: Steve, I believe you will find the answers to your questions if you re -read my replies to you.
Yes. That is a valid issue. No problem.
I simply stated that there are some things we can't control and some that we have some influence over. Do your best when you design.
Or, give up and built everything out of plastic or cardboard. This I did not suggest. It's just the other extreme.
So, the only thing I am wondering is why you singled my comments out. It does not appear that we are in disagreement at all.
-Chris
Edit: Steve, I believe you will find the answers to your questions if you re -read my replies to you.
Steve, cut it out.
Steve Eddy said:
Well, it's not an issue for me so I would have no suggestion along those lines.
I was just watching all this talk of how to make the chassis and couldn't help wondering if those who were so worried about shielding had given any thought to the cables that would be connected to it.
se
Just a thought if I may:
Cables are transmission lines where no amplification takes place and they are usually shielded; inside the case you have active unshielded devices and high impedance nodes where amplification takes place and many other unshielded components which of course are more susceptible to Rf and what not. This would certainly justify a good shielded case in my humble opinion. How good is another matter... but as good as possible.
CONSENSUS da Capo
Allright let's make amplifiers, down to earth, we need schematics of the JC-80 and a couple of good metal boxes, PCB and nuts and bolts and some voodoo but not so much just very little.
I'm a very bad DIY'er can't make any metal box or PCB but I can use the IRON.
Just open the workshops and begin to solder.
Kamskoma
Allright let's make amplifiers, down to earth, we need schematics of the JC-80 and a couple of good metal boxes, PCB and nuts and bolts and some voodoo but not so much just very little.
I'm a very bad DIY'er can't make any metal box or PCB but I can use the IRON.
Just open the workshops and begin to solder.
Kamskoma
anatech said:
That's fine. But not everyone's idea of "practical" nor how practical they choose to be will be the same as yours.
Some do, sure. But not everyone.
I guess you can't, and if one is interested in making such comparisons, they'll ultimately make different choices.
I guess it does. But you seem to say this almost as if being "safe from comparison" is somehow... cowardly or something.
That's fine. But unless one is a commercial manufacturer, I don't see that going one's own way is somehow illogical.
I can't say I see how it's necessarily shooting myself in the foot.
Not as such, no, but this is diyaudio.com, not commercialmanufacturers.com, yes?
Could be any number of reasons, not the least of which might be that one might achieve better results by going outside the ipso facto "standards" that commercial manufacturers have to adhere to if they expect to remain a commercial manufacturer.
Where I'm coming from is from an individual with my own particular tastes, preferences, and ideas as to how I might best satisfy my particular tastes and preferences. And being an individual and having only to worry about a market of one, at least as far as DIY goes, I see no compelling need to adhere to those standards which are required when one needs to worry about a market of millions.
I'm sorry to hear that.
se
apassgear said:
Sure.
Yes.
Fair point. But those same amplification circuits are still connected to the cables which feed into and out of the chassis. Not only that, the cables also allow for ingress into the chassis of RFI. So it would seem to me that one would want the shielding of the cables to be as robust as that they desire of the chassis.
se
Steve Eddy said:
Steve,
this is exaggerated. You need good shield (not necessarily 8mm thick) connected to box. Then RF filter placed directly at the input connector (in front of the inlet, placed in RF protected housing, like BNC - cylindrical housing - BNC) is the solution that works.
On must ensure continuity of shielding.
rdf said:
I don't know that that's necessarily the case. The loop areas of the preamp circuit can be made quite small if one is careful about layout and parts selection (such as surface mount devices). The twisted pair of wires may have a seemingly close spacing, but the length of the typical cable is many times greater than the size of the preamp circuit.
Also, far as I'm aware, loop area is mainly of concern as far as magnetic fields go which means you've still got electric fields to contend with.
se
PMA said:
Right.
And I just realized I wasn't thinking when I mentioned RFI.
The idea behind using as thick an aluminum enclosure as possible wasn't so much to shield against RF but rather to provide effective shielding against low frequency magnetic fields in order to avoid using much more effective but supposedly worse sounding ferrous materials.
se
Anatech and others,
The point I submitted about cables is that they can be real help or pure sake oil ! The dual cells pi-filter (Schaffner,....) just give an attenuation of 65 dB (yes, it is great, but depends on the attenuated band.). And about split-bobbin Xformer vs toroidals, I am very puzzled that Dick Sequerra USES a toroidal in his high end professional supply filter ! The converse John's own way, in fact... But the method used after it is perhaps very different to solve...
The problems, as I see them (short list)
- 60 Hz line (sorry, it is an USA figure…). You can put a 50 Hz here…,
- one is 440 Hz : industrial current spikes,
- one is laying from above 10 KHz to about 200 KHz with quite high spikes, and is more related to the switching power supplies (microvave oven supplies, computers, CRT monitors,…)
- above 500 KHz, there is the radio emissions (more likely old Russian emitters…),
- and above 1 MHz, it is a mix (low emitter subharmoncs, plane broadcast,…).
The protections (Surge , also short-list...)
- “Shunt-mode” suppression) : easily done in regard of transients by just shunting them to ground, with generic Metal Oxide Varistors (commercial names SIov (Siemens), or GEmov (General Electric)). BUT they are prone to fail themselves without a warning word, and just leaving the protected stuff alone. Also, they are life limited, and when they stop a surge, their own life reduces… They tend to pollute the ground by the way they act (shunt mode !), and are NOT protective for very high voltage spikes (lightning hits,…).
- “Series-mode suppression” : better, as they only store the surge energy in a capacitive way, and slowly release it into,
- Line filters : there are 2 types : “common mode” and “differential mode” coils. They are quite cheap and work quite good… But they ALL have the same problem : they are “directional”, and their own performance relies on both source and load impedances…
I just jumped from the outside from the case to the inside. But still clear to me that all these protections added just don't clear the EMI/RFI problem at such. I started another study to compare mains cables attenuation in regard of the graph I half drawn, and will report results... Then discussioncould occur. I was just pointing one possible direction here
Cheers aqll !
Jbaudiophile
The point I submitted about cables is that they can be real help or pure sake oil ! The dual cells pi-filter (Schaffner,....) just give an attenuation of 65 dB (yes, it is great, but depends on the attenuated band.). And about split-bobbin Xformer vs toroidals, I am very puzzled that Dick Sequerra USES a toroidal in his high end professional supply filter ! The converse John's own way, in fact... But the method used after it is perhaps very different to solve...
The problems, as I see them (short list)
- 60 Hz line (sorry, it is an USA figure…). You can put a 50 Hz here…,
- one is 440 Hz : industrial current spikes,
- one is laying from above 10 KHz to about 200 KHz with quite high spikes, and is more related to the switching power supplies (microvave oven supplies, computers, CRT monitors,…)
- above 500 KHz, there is the radio emissions (more likely old Russian emitters…),
- and above 1 MHz, it is a mix (low emitter subharmoncs, plane broadcast,…).
The protections (Surge , also short-list...)
- “Shunt-mode” suppression) : easily done in regard of transients by just shunting them to ground, with generic Metal Oxide Varistors (commercial names SIov (Siemens), or GEmov (General Electric)). BUT they are prone to fail themselves without a warning word, and just leaving the protected stuff alone. Also, they are life limited, and when they stop a surge, their own life reduces… They tend to pollute the ground by the way they act (shunt mode !), and are NOT protective for very high voltage spikes (lightning hits,…).
- “Series-mode suppression” : better, as they only store the surge energy in a capacitive way, and slowly release it into,
- Line filters : there are 2 types : “common mode” and “differential mode” coils. They are quite cheap and work quite good… But they ALL have the same problem : they are “directional”, and their own performance relies on both source and load impedances…
I just jumped from the outside from the case to the inside. But still clear to me that all these protections added just don't clear the EMI/RFI problem at such. I started another study to compare mains cables attenuation in regard of the graph I half drawn, and will report results... Then discussioncould occur. I was just pointing one possible direction here
Cheers aqll !
Jbaudiophile
Rdf, your input is right on. We don't really have much problem with the circuitry itself, BUT we do have a big problem with the Shallco switches and TKD pots as they are not well shielded and are very large in size.
Now why do we use these, rather than some teeny-tiny pot or switch, or relays, or solid state switches on the board itself?
Over the decades, and even today, I have used relays with cheaper products, and we board mount virtually everything that we can. Same designer, layout expert, and tweaker, (3 different persons). It just isn't as good as Shallco switches, but it's OK.
When we use these big, open frame switches, we get the best switching that we know about, but they are virtually perfect receiving antennas for RFI. That is why we try to make a sealed, thick walled aluminum enclosure. We don't like to use iron. Our major tweaker, Bob Crump, would never allow such a thing,
This is our choice, and we have found it to be a good one.
Now why do we use these, rather than some teeny-tiny pot or switch, or relays, or solid state switches on the board itself?
Over the decades, and even today, I have used relays with cheaper products, and we board mount virtually everything that we can. Same designer, layout expert, and tweaker, (3 different persons). It just isn't as good as Shallco switches, but it's OK.
When we use these big, open frame switches, we get the best switching that we know about, but they are virtually perfect receiving antennas for RFI. That is why we try to make a sealed, thick walled aluminum enclosure. We don't like to use iron. Our major tweaker, Bob Crump, would never allow such a thing,
This is our choice, and we have found it to be a good one.
Yes, John...
And it should have been asking Bob with a quite tiny voice, just ahead this gigantic guy...
Some of Bob Crump are here : "My business partner called earlier tonight and I explained what was going on with this thread
and he got a hoot out of it as he was pretty much at a loss to explain why the Blowtorch is so dynamic compared to most any other unit other than the topology is a complimentary differential
folded cascode with no loop feedback.
He used to use some 26 turn pots which had beaten up a unit he had designed in 1980, the Dennesen JC-80,
and then the Blowtorch came in and destroyed the wire wound pots.
Bottom line is there is more dynamic information with the CTC than there is with most anything out there. Curl made a joke and said it must be the atomic dynamicizer he designed into the circuit
Open loop bandwidth of 350K doesn't hurt either "
And it should have been asking Bob with a quite tiny voice, just ahead this gigantic guy...
Some of Bob Crump are here : "My business partner called earlier tonight and I explained what was going on with this thread
and he got a hoot out of it as he was pretty much at a loss to explain why the Blowtorch is so dynamic compared to most any other unit other than the topology is a complimentary differential
folded cascode with no loop feedback.
He used to use some 26 turn pots which had beaten up a unit he had designed in 1980, the Dennesen JC-80,
and then the Blowtorch came in and destroyed the wire wound pots.
Bottom line is there is more dynamic information with the CTC than there is with most anything out there. Curl made a joke and said it must be the atomic dynamicizer he designed into the circuit
Open loop bandwidth of 350K doesn't hurt either
"- Status
- Not open for further replies.