Elecraft

OT: The Great KPA Fan Mystery

Classic

List

Threaded

11 messages Options

Eric Norris-2

OT: The Great KPA Fan Mystery

Why does the KPA500 suck air from the top vents and blow it out the back,
and the KPA1500 sucks air from the back and blows it out the top vents?
Was there a coup in Elecraft engineering? Will future Elecraftologists be
looking for overwritten cartouches on the fan control boards? And who,
exactly, is buried in KV64?

73 Eric WD6DBM
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

ANDY DURBIN

OT: The Great KPA Fan Mystery

"Why does the KPA500 suck air from the top vents and blow it out the back"

I asked quite a while ago what thermal analysis had been performed on the KPA500 and got no answer here. It would seem reasonable for the fan to augment natural convection rather than oppose it.

Kenwood built a batch of TS-590SG with the fans installed backwards but eventually they corrected the mistake in production. I don't think Elecraft ever decided the KPA500 fan was the wrong way round.

Now for a KPA500 mystery question - does heat flow from the Z bracket into the heatsink or does it flow from the heatsink to the Z bracket?

73,
Andy, k3wyc
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

wayne burdick

Re: OT: The Great KPA Fan Mystery

Administrator

For each product, we experimented with air flow direction and many other variables. The current configurations produced the fastest and most consistent cooling.

73,
Wayne
N6KR

> On Apr 30, 2020, at 12:01 PM, Andy Durbin <[hidden email]> wrote:
>
> "Why does the KPA500 suck air from the top vents and blow it out the back"
>
> I asked quite a while ago what thermal analysis had been performed on the KPA500 and got no answer here. It would seem reasonable for the fan to augment natural convection rather than oppose it.
>
> Kenwood built a batch of TS-590SG with the fans installed backwards but eventually they corrected the mistake in production. I don't think Elecraft ever decided the KPA500 fan was the wrong way round.
>
> Now for a KPA500 mystery question - does heat flow from the Z bracket into the heatsink or does it flow from the heatsink to the Z bracket?
>
> 73,
> Andy, k3wyc
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]

NK7Z

Re: OT: The Great KPA Fan Mystery

I wish you all had included a filter slot... :)

73, and thanks,
Dave (NK7Z)
https://www.nk7z.net
ARRL Volunteer Examiner
ARRL Technical Specialist
ARRL Asst. Director, NW Division, Technical Resources

On 4/30/20 12:06 PM, Wayne Burdick wrote:

> For each product, we experimented with air flow direction and many other variables. The current configurations produced the fastest and most consistent cooling.
>
> 73,
> Wayne
> N6KR
>
>
>> On Apr 30, 2020, at 12:01 PM, Andy Durbin <[hidden email]> wrote:
>>
>> "Why does the KPA500 suck air from the top vents and blow it out the back"
>>
>> I asked quite a while ago what thermal analysis had been performed on the KPA500 and got no answer here. It would seem reasonable for the fan to augment natural convection rather than oppose it.
>>
>> Kenwood built a batch of TS-590SG with the fans installed backwards but eventually they corrected the mistake in production. I don't think Elecraft ever decided the KPA500 fan was the wrong way round.
>>
>> Now for a KPA500 mystery question - does heat flow from the Z bracket into the heatsink or does it flow from the heatsink to the Z bracket?
>>
>> 73,
>> Andy, k3wyc
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Lyn WØLEN

Re: OT: The Great KPA Fan Mystery

You just answered the question, Dave - and it's a two-fold solution for the
KPA500:

1) There is likely less dust on the top than in the rat's nest of wires and
cables found behind the unit (at least that's the case with mine, since it
sits within a small shelving unit), and
2) You can easily lay a filter pad on top of the KPA500.

Problem solved. You're welcome.

73
Lyn, W0LEN

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of Dave Cole
Sent: Thursday, April 30, 2020 2:40 PM
To: [hidden email]
Subject: Re: [Elecraft] OT: The Great KPA Fan Mystery

I wish you all had included a filter slot... :)

73, and thanks,
Dave (NK7Z)
https://www.nk7z.net
ARRL Volunteer Examiner
ARRL Technical Specialist
ARRL Asst. Director, NW Division, Technical Resources

On 4/30/20 12:06 PM, Wayne Burdick wrote:
> For each product, we experimented with air flow direction and many other
variables. The current configurations produced the fastest and most
consistent cooling.
>
> 73,
> Wayne
> N6KR
>
>
>> On Apr 30, 2020, at 12:01 PM, Andy Durbin <[hidden email]> wrote:
>>
>> "Why does the KPA500 suck air from the top vents and blow it out the
back"
>>
>> I asked quite a while ago what thermal analysis had been performed on the
KPA500 and got no answer here. It would seem reasonable for the fan to
augment natural convection rather than oppose it.
>>
>> Kenwood built a batch of TS-590SG with the fans installed backwards but
eventually they corrected the mistake in production. I don't think Elecraft
ever decided the KPA500 fan was the wrong way round.
>>
>> Now for a KPA500 mystery question - does heat flow from the Z bracket
into the heatsink or does it flow from the heatsink to the Z bracket?

>>
>> 73,
>> Andy, k3wyc
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Clay Autery-2

Re: OT: The Great KPA Fan Mystery

In reply to this post by wayne burdick

Wayne is correct.... (of course). Thermal performance VERY often runs
altogether in opposition to intuitive thinking on the application.

______________________
Clay Autery, KY5G
(318) 518-1389

On 04/30/20 14:06, Wayne Burdick wrote:

Clay Autery-2

Re: OT: The Great KPA Fan Mystery

In reply to this post by NK7Z

With a little searching online, you can find ultra-fine dust screens
that do not reduce volume too much.... that are also very thin, and can
be temp mounted on the intake vents without radically injuring the
aesthetics.

IF you care less about aesthetics and more about performance, I can
point you in the right direction to constructing an intake filter
assembly from OTS parts that will trap 90%+ of the pairborne
particulates with less than 3% volume and air velocity reduction.
Better than that IF you are willing to pay the prices...

BEST practices is to leave Wayne and Eric's creation alone, and simply
crack her open every 3 months and blow her out (CAREFULLY) with CLEAN
and DRY air...

73,

______________________
Clay Autery, KY5G
(318) 518-1389

On 04/30/20 14:40, Dave Cole wrote:

> I wish you all had included a filter slot... :)
>
> 73, and thanks,
> Dave (NK7Z)
> https://www.nk7z.net
> ARRL Volunteer Examiner
> ARRL Technical Specialist
> ARRL Asst. Director, NW Division, Technical Resources
>
>

Edward R Cole

Re: OT: The Great KPA Fan Mystery

In reply to this post by Eric Norris-2

My comment NOT about Elecraft cooling design, but:

I am finding a lot of amplifier designs place fans to exhaust heated
air with cool air being drawn into the heat sink by the partial
vacuum created by the fans.

My 6m converted Harris ch.2 TV linear is set up this way (amp has
openings in the chassis to circuits on input side as well as the
final transistors).

My W6PQL 1296 LDMOS 600w amp is cooled in the same fashion.

Only departure is my W6PQL 2m-1500w linear that has four 2-inch
square fans blowing cool air into the heat sink. But I augmented air
flow by adding a 4-inch exhaust fan on enclosure top cover. The
intake fans were separated from the heat sink by a 2-inch gap which
bothered me,so added sheet metal fairings to channel air to the heat
sink. Seems to work as heat sink temp remains topped out at 39c
running 1400w with JT65 (Overtemp trips off 50v supply at 55c).

Even my old 2m-8877 with blower had a 7-inch radial exhaust fan to
reduce back pressure on the blower.

73, Ed - KL7UW
http://www.kl7uw.com
Dubus-NA Business mail:
[hidden email]

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Christopher Hoover

Re: OT: The Great KPA Fan Mystery

I think we're getting far off the topic of Elecraft ....

The operating point for forced air cooling of a system should be the point
where the fan curve intercepts the system resistance curve.

The fan curve comes from the fan vendor.

The system resistance is measured on a flow bench in the lab, or you can
can use computational fluid dynamics. But the lab work is ideal if you
have the real thing or a decent mock up and is < 1 days worth of work).

The system resistance is not something you will typically have at hand
though unless you are the designer. Probably most amatuer designers never
bother. (It isn't very hard to make your own flow bench.)

You typically want to arrange the airflow within the system to maximize dT
between the hot things in the enclosure and the working fluid as the heat
transferred is proportional to dT (q' = h.dT). That's not always
possible for a variety of reasons. Things are in the wrong "order."
Sometimes certain components need to be kept cooler (e.g. camera sensor
suffers from unacceptable noise with Tj > 60C).

For a given fan type, flow rate and pressure drop, there is a fan size that
is maximally efficient and least noise. A smaller or larger fan cannot be
more efficient and hence less noise. I know some are likely to argue
about this. I refer you to [Handbook of Acoustical Measurements and Noise
Control, by C. Harris, 1991] for details. That said, you can't
(without paying a chunk of NRE) order up an arbitrary size fan, so there is
a compromise.

Adaptive fan control is another matter ....

73 de Ai6KG
-ch

On Fri, May 1, 2020 at 12:12 AM Edward R Cole <[hidden email]> wrote:

> My comment NOT about Elecraft cooling design, but:
>
> I am finding a lot of amplifier designs place fans to exhaust heated
> air with cool air being drawn into the heat sink by the partial
> vacuum created by the fans.
>
> My 6m converted Harris ch.2 TV linear is set up this way (amp has
> openings in the chassis to circuits on input side as well as the
> final transistors).
>
> My W6PQL 1296 LDMOS 600w amp is cooled in the same fashion.
>
> Only departure is my W6PQL 2m-1500w linear that has four 2-inch
> square fans blowing cool air into the heat sink. But I augmented air
> flow by adding a 4-inch exhaust fan on enclosure top cover. The
> intake fans were separated from the heat sink by a 2-inch gap which
> bothered me,so added sheet metal fairings to channel air to the heat
> sink. Seems to work as heat sink temp remains topped out at 39c
> running 1400w with JT65 (Overtemp trips off 50v supply at 55c).
>
> Even my old 2m-8877 with blower had a 7-inch radial exhaust fan to
> reduce back pressure on the blower.
>
> 73, Ed - KL7UW
> http://www.kl7uw.com
> Dubus-NA Business mail:
> [hidden email]
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Christopher Hoover

Re: OT: The Great KPA Fan Mystery

bad engrish there. it "will be" not "should be" 73

On Fri, May 1, 2020 at 3:08 PM Christopher Hoover <[hidden email]> wrote:

> I think we're getting far off the topic of Elecraft ....
>
>
> The operating point for forced air cooling of a system should be the point
> where the fan curve intercepts the system resistance curve.
>
> The fan curve comes from the fan vendor.
>
> The system resistance is measured on a flow bench in the lab, or you can
> can use computational fluid dynamics. But the lab work is ideal if you
> have the real thing or a decent mock up and is < 1 days worth of work).
>
> The system resistance is not something you will typically have at hand
> though unless you are the designer. Probably most amatuer designers never
> bother. (It isn't very hard to make your own flow bench.)
>
>
>
> You typically want to arrange the airflow within the system to maximize dT
> between the hot things in the enclosure and the working fluid as the heat
> transferred is proportional to dT (q' = h.dT). That's not always
> possible for a variety of reasons. Things are in the wrong "order."
> Sometimes certain components need to be kept cooler (e.g. camera sensor
> suffers from unacceptable noise with Tj > 60C).
>
>
> For a given fan type, flow rate and pressure drop, there is a fan size
> that is maximally efficient and least noise. A smaller or larger fan
> cannot be more efficient and hence less noise. I know some are likely to
> argue about this. I refer you to [Handbook of Acoustical Measurements and
> Noise Control, by C. Harris, 1991] for details. That said, you can't
> (without paying a chunk of NRE) order up an arbitrary size fan, so there is
> a compromise.
>
>
> Adaptive fan control is another matter ....
>
>
> 73 de Ai6KG
> -ch
>
>
>
> On Fri, May 1, 2020 at 12:12 AM Edward R Cole <[hidden email]> wrote:
>
>> My comment NOT about Elecraft cooling design, but:
>>
>> I am finding a lot of amplifier designs place fans to exhaust heated
>> air with cool air being drawn into the heat sink by the partial
>> vacuum created by the fans.
>>
>> My 6m converted Harris ch.2 TV linear is set up this way (amp has
>> openings in the chassis to circuits on input side as well as the
>> final transistors).
>>
>> My W6PQL 1296 LDMOS 600w amp is cooled in the same fashion.
>>
>> Only departure is my W6PQL 2m-1500w linear that has four 2-inch
>> square fans blowing cool air into the heat sink. But I augmented air
>> flow by adding a 4-inch exhaust fan on enclosure top cover. The
>> intake fans were separated from the heat sink by a 2-inch gap which
>> bothered me,so added sheet metal fairings to channel air to the heat
>> sink. Seems to work as heat sink temp remains topped out at 39c
>> running 1400w with JT65 (Overtemp trips off 50v supply at 55c).
>>
>> Even my old 2m-8877 with blower had a 7-inch radial exhaust fan to
>> reduce back pressure on the blower.
>>
>> 73, Ed - KL7UW
>> http://www.kl7uw.com
>> Dubus-NA Business mail:
>> [hidden email]
>>
>> ______________________________________________________________
>> Elecraft mailing list
>> Home: http://mailman.qth.net/mailman/listinfo/elecraft
>> Help: http://mailman.qth.net/mmfaq.htm
>> Post: mailto:[hidden email]
>>
>> This list hosted by: http://www.qsl.net
>> Please help support this email list: http://www.qsl.net/donate.html
>> Message delivered to [hidden email]
>>
>

ANDY DURBIN

Re: OT: The Great KPA Fan Mystery

In reply to this post by ANDY DURBIN

I don't think anyone was proposing taking on a new cooling design for the KPA500. The heat sink is there. The conduction path between the finals and the heat sink is what it is. The fan is selected and the fan speed temperatures are defined. The only speculation is whether the flow direction is optimum. Elecraft says it is.

I have over a year's worth of temperature data logged for my KPA500. My logger records, along with many other KPA500 parameters, the finals temperature, heat sink fin temperature, and Z bracket temperature, once per second for every transmission. I doubt anyone has a better set of temperature data for a KPA500.

If I decide to reverse the fan, I do have a good baseline data set against which to evaluate the change. However, the real reason for logging the temperatures was to establish a baseline against which to evaluate any mechanical changes made in an attempt to eliminate the annoying thermal clicks.

73,
Andy, k3wyc

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]