John Curl's Blowtorch preamplifier part II

Status
Not open for further replies.
SE,
Not so sure about whether the paper phenolic would be more hygroscopic than glass epoxy, but I will give you that the glass epoxy should be stronger in a shear application. Phenol resins are used in aerospace applications I believe due to the high heat distortion properties and resistance to burn, I think they char and somewhat self extinguish but it has been years since I talked to my friend who sold those chemicals. If I remember correctly the glass fibers in the trimmed edges of a glass/epoxy laminate are hygroscopic in nature. One of the failure modes of that material under certain conditions if again I am remembering

Yeah, but realistically, just how strong does it really need to be? Unless you're loading it with heavy transformers or big-*** caps and it needs to survive the rigors of shipping, it's a perfectly good material for audio. Also, FR1 and FR2, which are paper/phenolic based are fire resistant, though they're not the beautiful dark brown like the old XP and XXP materials.

se
 
SE,
I have a sheet of some of that in 1/4" thickness that I just so happened to use to mount some crossovers on, a bit of overkill I would have to say. Looks much more like standard pcb board color though with a slight greenish tint.

Back to audio for now and trouble shooting a car problem no mechanic can find. Of course it is an electrical problem, intermittent at that, heat related under the hood.
 
Yeah, I use those materials. Also Solvay and Arkema.

Some are so soft they can be stripped from the wire with a fingernail if not too thick...

Nice link, thanks.

Their velocity equations are valid only for coaxial cables as well as for striplines as long as the aspect ratio is large, IE very thin insulation with conductors at least 10 to 100 times wider. Once we go to parallel runs where the magnetic and electric fields are no longer constrained, we have to modify the equation a tad..v = 1/sqr(LC), or 1/sqr(EDC), or 1/sqr(epsilon*mu).


jn

My pleasure. How about calculating an ~9awg star quad (12awg per conductor) that is moderately constrained (0.030 between conductors, Teflon insualted) and 5-6 twists per foot?

Yes. The problem is in the level of inductance. Cables run the range from 200 nH per foot down to under 10nH per foot, depending on design. Many vendors do not auto-zero their instrument or realize it is needed, some will use Lp/Rp instead of Ls/Rs, some will try to measure 100 nH using 100hz or 120 hz test frequencies, and many do not rigidly control the setup to prevent variation of parasitic inductance.

Capacitance turns out to be significantly easier to get valid or near valid values close to reality.

jn

Using a GR 1689M LCR meter down near the bottom end of the Ls range is challenging. To get repeatable numbers I use a 4 wire - to - 2 wire fixture and have to re-zero at each new frequency.

By "parasitic inductance" are you referring to proximity effects from things like metal benchs, loops made in the cable under test, etc.?

As an aside, do cable AC parameters change appreciably at higher currents other than IR losses? Instead of testing cables at 1mA or 10mA for example, what happens with the increased fields at 1A? 10A? higher?

Thanks,
Dave
 
FYI,
PCB substrates used 2010, figures have not changes much in last couple of years. Both the HDI and Std. multilayer are FR4 based
 

Attachments

  • PCB_Production.jpg
    PCB_Production.jpg
    153.5 KB · Views: 213
AX tech editor
Joined 2002
Paid Member
I know it's the brown very breakable material
used for boards in the cheapest devices.
As opposed to fibreglass used in more quality
conscious equipment.

No that's not Bakelite. That's phenolic board, as explained by SY. Bakelite is a different material - I believe, for instance used in the past for light switches, kettle handles, that sort of thing.

Jan
 
Real Bakelite can be very beautiful, and in the '50's was often swirled of two colors for consumer goods. It can be cast very smooth and take detailed shapes, like some Kodak cameras and lots of "Art Deco" radios. Hand-held 35mm slide viewers are sometimes magnificent little jewels. Beautiful stuff of an earlier age; too expensive nowdays.

Find a 1950's, early '60's, Leitz Wetzlar slide projector and look at its detachable power cord's connectors, with integral switch, and enjoy real engineering beauty.

Thanks,
Chris
 
Member
Joined 2002
Paid Member
Many vendors do not auto-zero their instrument or realize it is needed, some will use Lp/Rp instead of Ls/Rs, some will try to measure 100 nH using 100hz or 120 hz test frequencies, and many do not rigidly control the setup to prevent variation of parasitic inductance.

Capacitance turns out to be significantly easier to get valid or near valid values close to reality.

It gets worse (or is it better? :D). All cheap hand held LC measuring instruments I have used do not test with a sinusoidal signal but with a pulsed, irregular squarewave one.



Using a GR 1689M LCR meter down near the bottom end of the Ls range is challenging. To get repeatable numbers I use a 4 wire - to - 2 wire fixture and have to re-zero at each new frequency.

This is my experience with cheap LC meters too.
For L measurements at the low scales on cables, I get a rather broad range of readings say +/- 10-20% of the calculated expected value and this with only some instruments, while for C measurements the range is much narrower~ +/-2-3% with all instruments.

By "parasitic inductance" are you referring to proximity effects from things like metal benchs, loops made in the cable under test, etc.?
To quote:
"inductance can only be properly understood if it is realised that it is only defined for a closed circuit.
What is usually meant by a statement about inductance is incremental inductance, such as that
caused by a length of wire if it were to be inserted into a closed circuit."

As an aside, do cable AC parameters change appreciably at higher currents other than IR losses? Instead of testing cables at 1mA or 10mA for example, what happens with the increased fields at 1A? 10A? higher?

Jn will come and provide his valuable comment on this.
My side-comments: If higher current increases the network losses, measurement system’s noise figure will worsen.
I want to hint on the –measuring instrument dependent-risk that the resulting higher voltages at the input of the instrument may increase intermodulation distortion products. This effect is not as easily verified as the effect of input overloading, which may also occur.


Steve, it is because of the constant level signal during the test. No real music will stress amplifier at constant 1/3 of rated power during 30 minutes.

Cube test (high amplitude pulse sweep excitation) is a realistic hard test, no?


Real Bakelite can be very beautiful, … like some Kodak cameras and lots of "Art Deco" radios.

Not to mention some TT-arm beauties :note:

George
 
Member
Joined 2002
Paid Member
There is a significant expense associated with them. Likewise, the hack designers that are common in fashion audio wouldn't know how to use them properly.

So true. An autotransformer, like a typical OT transformer, is a matching impedance device.
In addition to this, it is a delicate to tune L/C LP filter for HF signal rejection (HF generated either at the amp or at the load side).
Tuning the transformer with a capacitor that shunts the primary or secondary winding (taps for an autotransformer) can make it.
The inductance of each winding (taps) is the decisive element in that tuning .

George
 
Steve, it is because of the constant level signal during the test. No real music will stress amplifier at constant 1/3 of rated power during 30 minutes.
With modern hypercompressed music releases that also have heavy low bass content that's easy. Think some techno, speedmetal and similar. Kids don't listen that audiophile 'cr*p', they want REAL music. Throw a loud party while daddy is not home and watch his posh amp go up in smoke :D
 
With modern hypercompressed music releases that also have heavy low bass content that's easy. Think some techno, speedmetal and similar. Kids don't listen that audiophile 'cr*p', they want REAL music. Throw a loud party while daddy is not home and watch his posh amp go up in smoke :D

Another reason to have good protection circuitry like Mac does. My kid is scared to death to touch Dad's stereo, which is also effective protection.
 
Status
Not open for further replies.