John Curl's Blowtorch preamplifier part III

Status
Not open for further replies.
Jump to Latest
AKN

AKN

Member
Joined 2005
Paid Member
Hi,

PMA is correct.
RF needs to be shunted to chassis, with small capacitance (not affecting audio) between input/output GND connectors and chassis.

Lots of RF pollution today. Think Wi-Fi Mobile, phones ETC.

My favorite scenario is High End exhibitions where lots of people (with phones) sits tightly together and do 'serious listening'. All mobiles are fighting for connection and increases RF level in unison.
 
D

DPH

Member
Joined 2008
billshurv said:
John, since when has best practice EMC been 'compromising' a product? Serious question as a roll off above 250kHz surely cannot cause any issues other than in the minds of the reviewers?
Click to expand...

Well, that goes a heck of a lot further than just 250kHz audibiliity, but we're already far past offending certain folks sensibilities with belt/suspenders engineering. ;)

Also, BT has long since sold and any newer design requires, theoretically, emc compliance. So no use flogging the bt.
 
Last edited:
C

chris719

Member
Joined 2004
PMA said:
Again, misunderstanding or misinterpretation, or both. Intentionally or not, who knows.
I have never said that thick aluminium case had potential flaws in picking up RFI. The case itself is OK and is very effective as a shielding. But, the wire input through insulated RCA connectors is a big, big problem (please see the CTC box and RCA connectors on the left and also next to XLRs). This is an input gate for RFI arriving in along the signal wires. Measurements to be done to check.
I am attaching measurements of a similar box with insulated RCA connectors and link preamp inside. Measured with Agilent RF spectral analyzer. One can see that up to some 20MHz HF attenuation is good. But then it starts to have wild behavior, as seen from measurements up to 100MHz and 1GHz. Input on-board RC filter does not help, the HF signal, once allowed to get in the box, easily bypasses audio circuits and appears at the output attenuated not very effectively.

In case that you have different results, please show them. Your customers would never know this and would have no chance to make a comparison with proper solution. No wonder they do not complain.
Click to expand...

With unbalanced connections it's always a bit of a compromise. I think I have seen some commercial Class I gear with chassis-connected RCA connectors that make their connection between chassis and circuit reference at the input connectors.
 
john curl

john curl

Member
Joined 2003
This reminds me of 1959 when my orientation professor in college defined a pseudo-intellectual with a story: "He was at a cocktail party when he asked for his martini extra 'dry', and someone came up an asked him to define 'dry'." (not the bartender) This 'better' question reminded me of this, sorry if I went overboard.
 
john curl

john curl

Member
Joined 2003
I like RCA connectors. Quality RCA's like we use are convenient, and 'universal'. Mark Levinson was one of the first people in the USA to deviate from RCA connectors back in the middle 70's for Fischer or perhaps Lemo connectors. All this did was force people to use Mark's cables or to go through an additional RCA to Fischer connector. How inconvenient, expensive and potentially compromising! DIN connectors are just as bad.
 
D

DPH

Member
Joined 2008
ridikas said:
I think John is in the business to make something sound better, not worse...
Click to expand...

Like something that doesn't let a huge amount of RFI ingress and risk of rectifying into audio frequencies? It's best practice to not let such RFI into your beautifully machined chassis than to filter (but, both in reality). Sure he uses a lot of current and FETs, which mitigates against the issue. But it's a real issue nonetheless.

http://www.analog.com/media/en/training-seminars/tutorials/MT-096.pdf

Anyhow, RFI is an issue that supersedes just the Blowtorch, so it's good to pay attention to what's being said about good engineering to get the best out of one's circuit than (clearly blind) adulation.
 
john curl

john curl

Member
Joined 2003
I have had RFI issues in my lifetime, but usually not in my own designs. Once, I was living in 'line of sight' of the main TV tower in San Francisco. I had to change my cables to foil shielded to remove the RFI that was effecting my Dyna PAS3 tube preamp. I haven't had trouble like that since. However, in the '50's and '60's we had a huge radar installation in the Santa Cruz mountains that would put out a pulse every 15 seconds or so, and it would overload just about everything. It's gone now thank goodness. That would have been a challenge to engineer a fix. '-)
 
Status
Not open for further replies.
Top Bottom