Hi Chuck,

I found the device I was thinking of. It's an Analog Devices ADP3000AR-5. Here's the link --> http://www.analog.com/UploadedFiles/Data_Sheets/ADP3000.pdf

Its operating temp range is -40C to +85C in an 8-lead SOIC style package.

Here’s the block diagram:

Unlike most of these step-down converters, this one switches around 400 kHz regardless of the load. A wider temp range would be nice. I’ll keep looking...

73, Juan

-----Original Message-----
From: Chuck Green [mailto:greencl@mindspring.com]
Sent: Sunday, July 29, 2007 7:28 AM
To: juan-rivera@sbcglobal.net
Cc: 'Bdale Garbee'; 'David Smith'; 'Dave Black (Work)'; 'Dave Black (Home)'; eagle@amsat.org; 'Samsonoff@Mac. Com'; 'Juan.Rivera (Work)'
Subject: Re: [eagle] Re: Receiver Spec vs. ATP, a few Suggestions and aQuestion or Two

Hi Juan,

My use of the value 90% (efficiency) is based on your measurements of

current supplied and my previous measurements of power consumed. But it

really doesn't matter how good this figure is, I think we can do a

little better.

I really suspect the existing regulator is running normally. It is

likely in a very low power mode and therefor not running at the 100 kHz

+ speed it can run at. But I'm not defending it. If there is something

better (more suitable for our current needs), great! You may very well

have found such a device. However, since it would be running at 10-15%

of capacity it may also switch to a more efficient mode (slower

switching speed).

Given the switching speed the regulator is running at now, I doubt the

layout has much to do with the issues you have uncovered. But that

really doesn't matter because we certainly wouldn't design a new board

to install the device dead-bug stile. However, you might want to

experiment with the new device as a direct replacement on the board

which would likely mean dead-bug stile. The topology of switching

supplies are often very much alike so it may well be that you could do

this as a first experiment.

And I wouldn't give up on your idea of an analog supply. Just because

one or two of us are not excited about this doesn't mean it isn't the

right thing to do. It's a system level issue of power management. But

I still want to encourage experimenting with a new switching supply to

see if it can resolve the problem.

Chuck

Juan Rivera wrote:

> Hi Chuck,

> I don't think that existing supply is running at 90% efficiency. I'll have

> to solder it back together to take a few measurements but the manufacturer's

> specs don't claim quite that much and this one isn't running properly. I've

> got a bit more data in my symposium presentation and I hope to complete that

> in the next few days and get it out.

> I found a step-down converter that's designed to supply a maximum of about

> 100 mA and it runs at 400 kHz with automotive temp range. If no one else

> steps up I'll put out some info on this one.

> 73,

> Juan

> -----Original Message-----

> From: Chuck Green [mailto:greencl@mindspring.com]

> Sent: Saturday, July 28, 2007 3:30 PM

> To: juan-rivera@sbcglobal.net

> Cc: 'Bdale Garbee'; 'David Smith'; 'Dave Black (Work)'; 'Dave Black (Home)';

> eagle@amsat.org; 'Samsonoff@Mac. Com'; 'Juan.Rivera (Work)'

> Subject: Re: [eagle] Re: Receiver Spec vs. ATP, a few Suggestions and

> aQuestion or Two

> Hi Juan,

> I think the reason you have not seen much regarding the CAN-Do redesign

> is that, as far as I know, no one has stepped up to do the power supply

> redesign. Until that happens, not much else will happen.

> I'm not thrilled with the idea of giving up the switching power supply.

> Remember that the efficiency hit must be multiplied by 10-15, the number

> of CAN-Do's on the satellite.

> The "dead-bug" modification is intended to fly. Those who were not

> around in the early days have no idea what loose wires have flown

> successfully in the past. That's not to say we should encourage such

> things, just that it's not as ugly to those of us who have seen much

> worse in the past as you might think. Because of the IC's mass (very

> small) it is probably very secure just the way you see it. But we will

> also epoxy a radiation shield to it and then conformal coat the whole

> thing with a heave conformal coating which will make it very difficult

> for anything to move.

> I'm ready to start a new design (layout) just as soon as someone gives

> us a new power supply design that is quieter, and hopefully, more

> efficient than the 90% we now have.

> Chuck

> Juan Rivera wrote:

>> Bdale,

>> It would be a trade-off. I've put out all the information I have and

>> everyone knows my opinions. I think someone else is supposed to be

> looking

>> into this but I forgot who it is since nothing has been posted. I'd like

> to

>> see some alternative suggestions from the experts on the CAN-Do, the

>> enclosure, and the EMI situation in general.

>> By the way, do I have a prototype CAN-Do module or was the intent to fly

>> this version with the dead-bug step-down converter hanging by three leads

>> and a few wires?

>> Juan

>> -----Original Message-----

>> From: Bdale Garbee [mailto:bdale@gag.com]

>> Sent: Saturday, July 28, 2007 2:41 PM

>> To: juan-rivera@sbcglobal.net

>> Cc: 'John B. Stephensen'; David Smith; Dave Black (Work); Dave Black

> (Home);

>> eagle@amsat.org; Samsonoff@Mac. Com; Juan.Rivera (Work)

>> Subject: Re: [eagle] Receiver Spec vs. ATP, a few Suggestions and

> aQuestion

>> or Two

>> On Fri, 2007-07-27 at 21:23 -0700, Juan Rivera wrote:

>>> I have a few thoughts... The CAN-Do switching step-down converter is

>>> only supplying 11 milliamps. If we take a slight efficiency hit we

>>> could just go to a simple linear regulator and completely eliminate

>>> the radiated and conducted EMI emission problem from CAN-do. That

>>> eases the EMI filtering and shielding requirements for every single

>>> payload. That seems like a good trade-off to me.

>>>

>> Hrm. What makes you say "a slight efficiency hit"?

>> Doing this on one or two modules that are particularly susceptible to

>> noise *may* make sense (and I'm certainly open to considering this as an

>> alternative), but we're already on our second power supply design on the

>> CAN-Do! because the original switcher, which was more efficient than a

>> linear regulator, was deemed too inefficient to fly on P3E by our

>> AMSAT-DL friends.

>> Bdale

>> _______________________________________________

>> Via the Eagle mailing list courtesy of AMSAT-NA

>> Eagle@amsat.org

>> http://amsat.org/mailman/listinfo/eagle