maybe "lots of improvement" would help... there are many people here suffering
im shocked it interferes with GPS...sice GPS is much higher freq
mickz, are you talking about: http://www.jaycar.us/productView.asp?ID=XC4876
also, can you post any images from your test equptment showing the noise?
i have no idea how to get this to the m4 people
im only interested case i do the HD Radio interfaces for carpcs here
(and sirius, and xm...)
anyone try the power supplies from say mini-box.com?
the design is quite different
Yes, thatís the unit. Only problem is it wonít bring my PC out of suspend.
However some of these things are MB specific. The M4 had no problems there.
Itís nothing fancy as there is no ignition auto start, easy to add something though.
There are surface mound devices on the underside of the board. Itís nothing flash
but itís a hell of lot quieter than the M4 and it works.
If I can find out why it fails to bring the MB out of Suspend to Ram Iíll knock up
a micro to control it correctly- unless someone beats me to it.
I'll see how I go getting something from the Spec An, however I think youíre wasting
your time with that; they will put it down to a faulty unit and not representative of product.
Like I said it desensitises an external mounted GPS and almost completely blocks anything mounted in the car
Just a quick note and a warning on the Jaycar unit.
This unit produces less noise in part because it does not boost the 12v rail.
This is the circuit that causes most noise in the system.
The 12v rail is not regulated. This will make the unit unsuitable for most automotive
installations. Regulating the 12v rail would be a simple task because of the low regulator
voltage drop involved. However it may not be worth the trouble for the reasons stated.
It works in my installation as I have a separate auto-isolated 12AH battery to power the PC,
so Iím not worried about cranking voltages, however the 12v output will rise to 13.4 or whatever you alternator charges at.
Itís a temp quick fix for my installation until I get something sorted out once I have
a base noise floor that I can tolerate and work from.
I have made the M4-ATX very, very quite – Virtually nil interference.
I decided to give this sh%$ of a thing one more go and fit it into an RF tight
case with feed through caps on all leads.
On the bench with a load I could sit the radio on a weak station on top of the PSU with
barely any change to reception. Previous to this you could not walk in the room with the radio.
Anyone interested in this mod could makes a case out of copper clad PCB.
Just cut and solder all internal seams, fit a dozen .001 feed through caps and
Remember these pictures are of a knocked together prototype to see if it would work
and produce enough RF attenuation – It does!
Now to make a case a bit smaller.
BTW, it took me over 2 hours of tearing my hair out to find out why the M4 failed to work
correctly after putting it in the case.
It would power up in dumb mode with 12v and nothing else. No 3.3v or 5v.
Got out a huge magnifier and went over the board looking for the usual suspects, and just when I was about to junk it,
I noticed a resistor position marked on the PCB, but no resistor. I could just see where one had been soldered.
Dam surface mount resistor had been knocked off the #$%^&* board.
Put a pot in circuit and wound it down until the 3 and 5 volt regs started. - Sigh of relief.
This resistor appears to control the drop out voltage and current limit.
If someone would be so kind as to measure the one on the PCB I would be greatly appreciative,
I would like to have the correct value in there. At the moment I settled on a 330ohm.
The resistor is on the edge of the board between J6 and the mounting screw.
The SMD resistor marked R2? I busted mine out of the dusty drawer as I wasn't doin' anything, reads as 6ohms, not k, just 6 flat. Keep in mind this is in the "'aw hell nah I ain't usin' that" drawer for a reason.
Originally Posted by Mickz
Quick iphone snap below
Thanks for taking the time to measure that, however the two boards are different.
Mine is the latest release and the resistor connects to C7. Unfortunately I canít see the
resistor position reference as itís hidden by my temp fix, which Iím loath to move at the moment.
Yeah, didn't look closely before at your snaps. I did buy mine quite a long while back now.
Originally Posted by Mickz
Do you think that this would work reasonably well with just using the PCB walls, without adding the caps to the lines?
Originally Posted by Mickz
I'm interested in doing this for my M4, as well as my SSD (Might make a dual chamber one for both of them to fit), but not sure I want to cut my cables to add the caps.
Otherwise, I'm happy to build one and test for an improvement!! heheh
The feed-through caps are absolutely vital.
You can leave the lid off the box and it is still fairly free of interference.
Just one lead sneaked over the edge of the case and onto the IGN connection was enough to
start blocking the radio and causing RF leakage.
This way of eliminating or reducing noise is standard RF practise.
You will need a few ventilation holes in the case, 4 or 5 Ĺ inch holes are fine, and a small fan
should be included as air flow is restricted.
The problem is RF crawling along the leads Ė nothing will remove this interference except a
correctly screened and bypassed lead entry/exit point. The case and feed through caps
Radiation from the PCB is minimal, itís the leads. External choking and bypass will fail.
Only this construction will work correctly. (For the purpose of this simple bypassing scheme)
These caps should be available from Amateur radio (Ham) parts outlet.
There are two types of mounting construction for feed-through caps.
One you solder to the case after pushing it through the hole drilled in the case, and the one
you bolt into the hole in the case.
You will notice that not every lead on the plug needs a feed-through. All the +5 volt leads are
joined on either side of the feed-through, the 3.3 and 12v leads do the same. This leaves a
few control lines and the + 12v and IGN to the PSU.
The USB interface is something I will look at tomorrow, as data lines, unless very low
impedance, do not take kindly to being bypassed with a cap.
I had two junked PC ATX supplies; I just cut the ATX leads off and used one inside the box
and one to the MB. These should be a dime a dozen.
Ok, I get the idea of the caps now, and from trawling mouser etc, they look easy enough to fit!
This gives me a good excuse to rebuild that part of my carpc, I'll have to pick up some parts and have fun soldering again!