Minimum heat-sinking requirements? (completely outside of the metal case)

[Ronald Nicholson]

If one were to operate a HL2 at 5W outside of the recommended 40mm or 55mm cases (for display purposes, scope probing both sides of the PCB, or ultra-lite SOTA/portable ops), what might be the recommended heat sinking?  

Has anyone though about or tried a plastic or plexiglass enclosure? (with some sort of heat sink).

Thanks,

Ron

n6ywu

[Steve Haynal]

Hi Ron and Group,

Some people do heat sink the HL2 in other ways. There is a large exposed area on the bottom of the PCB seen in the picture below. The game is to pipe heat from that area off of the HL2. The HL2 does have temperature monitoring which will disable TX if the temperature is >55C, so you should be safe with experimenting. I like to see the HL2 at below 45C on most days, and below 50C on the hottest days.

I would like to see a small aluminum rectangle manufactured that is about the width and length of the exposed area in the picture below, with hole for the screw, and thickness that matches the distance between the bottom of PCB and enclosure surface when the HL2 is mounted in the recommended enclosure. Such a heat shim could be used in the standard enclosure to provide better thermal transfer, or for those who want to use other finned heat sinks as it would provide a spacer so that the bottom of the HL2 PCB is not shorted by the heat sink. Makerfabs does do custom CNC work, and I have HL2 credit with Makerfabs. If someone is willing to draw up such a heat shim in one of the formats accepted by Makerfabs, I can have a bunch made. A free shim goes to the person who draws it up properly.

https://www.makerfabs.com/cnc-service.html

73,

Steve

kf7o

[Jayson Bucknell]

Steve, I've been looking for one of those shims ever since I noticed M1GEO's fine example in copper.

https://twitter.com/m1geo/status/1281953922672984066

Asking him the measurements he basically described them as 27 x 6.2 x 5 mm With the 5mm dimension as the z-axis, approximately 3.3mm hole centred through the middle of the z-axis for the screw between the board and the case. Fly cut for a smooth finish.

I believe this was for the standard case with the kit. Not sure about the creating the model, but if there'a batch made i'm interested.

73,

Jayson

AA7NM

[Steve Haynal]

Hi Jayson,

Thanks for the link. Yes, that is exactly what I'd like to make a batch of. It is a very simple shim and just needs to be in one of the accepted file formats:

AMF, CGR, EPRT, HCG, HSF, IFC, IGS, IGES, PRT, PRTDOT, SAT, SLDLFP, SLDPRT, STP, STEP, VDA, WRL, XAML, X_B, X_T, 3DXML and 3D DWG file types for quotation.

I can get quotes for aluminum or copper.

73,

Steve

kf7o

[Chris Gerber]

Hi

Enclosed a drawing of a cooler parrt I have mounted here to and some other HL2

Maybe someone can test it and see if its suitable. Here it works as drawn

I have sent the drawing some weeks ago to Makerfabs and they aswerd to

forward it to to Steve to look at it.

73 Chris Hb9bdm

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/756a66b2-348c-4125-8cff-1f2f539eec81o%40googlegroups.com.

[Steve Haynal]

Hi Chris,

That looks like a nice drawing. Someone just needs to port it to one of the formats accepted by Makerfabs. Any volunteers?

73,

Steve

kf7o

[samuel kallmeyer]

Hi Steve & all,

Find attached the freecad project with the associated .step (and .stl in case of) file based on Chris drawing.

If it can help and if someone is able to double-check before sending it to Makerfabs, feel free to use it ;)

73,

Samuel

f8acb

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/8bfb449e-adc5-49c8-8f60-b2360d6739ban%40googlegroups.com.

[Steve Haynal]

Hi Samuel,

Thanks for doing this. I am on vacation now but will review and hopefully submit to Makerfabs within the next week.

73,

Steve

kf7o

[Chris Gerber]

Hi Steve, Samuel

The pieces working great here but I have noticed that some boards have a rather large ground solder joint at one corner of the

cooling parts side at the reseved part so it should be checked.

If you like I could make a new drawing with a correction as have two boards here to check. So we are on the safe side.

It is also important to use the correct screw size of 3mm so its kept on the right place. A smaller screew size could missplace it

Chris

hb9bdm

[k7ul....@gmail.com]

I attached a 5mm square chunk of keystock by carfully drilling a hole in it.  Coated it with compound as not react with the copper or aluminum case.  I was worried about board deflection while tightening it down, so that was my best option at the time.....

I am eager to see what better solution becomes available.

73, Andy

[Steve Haynal]

Hi Chris and Group,

That would be great if you want to alter the drawing to avoid places where solder might be. 

For those who may be casually reading this thread, I believe the the current recommended enclosure with heat transfer via the slot contact is sufficient. See assembly procedure. A heat shim like we are discussing in this thread is extra, and might be useful for those operating in very hot ambient conditions, those wishing to use other types of thermal dissipation, or those wishing to increase the power output. Also, there is some credit to spend with Makerfabs and I'd like to try their CNC services. Peronsally, I am interested in eventually trying a different LDMOS device to increase the power output to 10W and think I will need better thermal dissipation for this.

73,

Steve

kf7o

[Chris Gerber]

Hi Steve

Here a corrected version of my cooling device. Its now for mountig at the safe side

73 Chris HB9BDM

[Doug Schultz]

If someone can machine them, I would buy one for sure ...

It seems much better than trying to get the heat out along the rails.

Doug

[Jayson Bucknell]

I'm not sure what normal operational variance is, but in my case transmitting causes an rapid rise in temperature over 20 degrees C to the low 50's. Even idle current in SSB with no modulation gets very hot. I've tried hacking up a rpi heatsink as an additional radiator on the screw but it doesn't help much. Cool down is rapid but decreases with the duty cycle. The edge connection just doesn't seem enough.

I've had enough little LDMOS die from thermal runaway to be paranoid. Even if the finals can handle it with the thermal protection the longer term rapid flexing of the circuits doesn't sound good.

If there's a design worked out I'd recommend it as part of the standard kit. I'm not going to try much in the way of WSPR of SSTV until then. It's been a great radio otherwise.

73,

Jayson
AA7NM

[dick_...@hotmail.com]

I think there are two different trains of thought here.  The subject is "(completely outside the  metal case)" and my trusty ruler indicates that there is only about 3.8mm of depth between the PCB and the bottom of the case [not 5mm].  Also the "3mm" hole is usually tangential to the screw boss that runs the entire length of the case, so I am not certain of the need for the 45 degree chamfer as  this would never fit in the normal case with out some serious rework along with reducing the height from 5mm to approx 3.8mm.  Conclusion is this part is ONLY for mounting outside the case.

Attached are 2 pix one shows the approximate location of the part on the PCB [without the chanfered edge] and the second shows what I think would be a much more desirable size that would actually be under the two fets and provide more thermal transfer.  As long as the part is for mounting outside the case it could be made larger and thicker to accommodate various heatsinking variations.

No one  has commented on the need to polish both the top and bottom to get the heatsink to work properly.  Online Metals sells copper bar in 1 foot lengths at various thicknesses.  However my experience with any copper or aluminum parts, is they arrive in various stages of oxidation the need attention.

Since I do not have a milling machine, I too would be interested in one that would fit in the case.  For outside the case I would use 1.1" x 1" x 0.25" stock that I have and just drill a hole, for a #4 or M3 screw, in the proper location and polish up the top and bottom.

Dick K9IVB

[Chris Gerber]

HI

Wrong, it is as stated 5mm (Measured and Tested) and so installed in already 4- 6 new Makerfabs 45mm cases.

Piece made from Alu, Copper would be better!

It extends the time to reach 50C Temp. from 8 Min to 23 Min. (Normal install,  versus Cooler piece istalled).

TX Datas: Power 5.6W, Current 1.4.6A at 13.6V on 14Mhz

Checked on my two HL2 units, both identical. The shape is simple to make and gives a real good heath transfer from board to case.

If some one has a better idea, fine with me, I am not in a competition.

73 Chris HB9BDM

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/4415cbf3-e522-42c5-9d79-1d08b13b850dn%40googlegroups.com.

[dick_...@hotmail.com]

Not necessarily wrong, as I have  Build 8 and Build 9 boards with a 106 x 55 x 100 mm case.  Apparently the new 40 x 105 x 100mm case shipped with the recent orders has different dimensions.  Upon reviewing the case info that I have for my cases and a lot of previous build cases I find two dimensions for the space from the bottom of the case to the bottom of the pcb.  These dimensions are 3.5mm and 3.6mm which are  somewhat less than my ruler measured but no where near 5mm.  There must be differences in the 40mm case and the 55mm cases.

If some one that received the wrong faceplates [55mm] and now has the new ones for the 40mm case were to try them on the ends there should be a significant difference in fit between the two faceplates, - just as a check.

Before anyone jumps in to get a spacer it would behoove them to start with a ruler.

Dick K9IVB

[samuel kallmeyer]

In case the corrected design from Chris would fit, I updated the files accordingly.

Find them attached.

Samuel, f8acb

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/4793edac-2111-43b4-965f-410b64894747n%40googlegroups.com.

[DL1YCF]

I made the same observation. Upon TXing in FT8, the temperature (as displayed by pihpsdr) rises by about 1 degree
per second, but also comes down nearly as quick upon RX.
I certainly would not do JT65 with 5 Watts. However, this was a hot day and the case
temp. was between 25 and 30 degrees Centigrade.

> --
> You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/3a668969-cd9f-45a7-adaa-eca0b5b6faabo%40googlegroups.com.

[Paul Keller]

 I confirm that the dimensions Chris communicated are correct. At least for the last version of cases I received.

73's de Paul HB9AXA

PS. It is not nessesary to dremel out the part of the side wall of the case as described in earlier documents. Just sand the bottom in order to make good contact.  

[Chris Gerber]

Hi Samuel

The size is checked and correct for the 45mm case, the once chipped at present. I know of 6 such cases which had all the dimensions

of 5mm. If there are different sizes around, this is not my concern, as I did it for the current make. All those mixup is more then confusing, and should be checked before hand.

At this place I would like to thank Steve for his fantastic design of the present HL2 units fabricated by Makerfabs.

A real great job, I had many SDR's with low power final stages in my hands, but your one is one of the best, and withstand even rough

conditions without the cooler.

But as you know, how lower the temp. the better, as my test approved. (I use it mainly for high duty cycle Digi modes)

So tought the cooler could be a nice ADD ON  for Hermes Lite 2 - to keep it cool, as well as heath buffer.

No solidstate device likes breathing, when used on intermittend modes. A constant kept temp. is very healthy.

This concludes my comments on the subject.

Best 73 Chris HB9BDM

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAFRyv0KUuz3g-h%3DrN-HT2ZbYOrJvdR9V61wafuUD9t1bNO7%3DcA%40mail.gmail.com.

[Steve Haynal]

Hi All,

I think the difference between 5mm and 3.5mm clearance must be between the 40mm versus the 55mm tall enclosure. I only have 55mm enclosures currently and measure 3.5mm with my calipers. I will have to order one of the 40mm tall enclosures. It would have been nice if the clearance was the same for both.

We actually did quite a bit of testing to conclude that dissipating heat via the rail was sufficient and economical. For example, a key down at full power for 10 minutes (actually 5 back to back 5W WSPR transmissions) did not exceed 50C when operating in typical ambient temperature of ~25C. The temperature will rise quickly during initial TX, but the rise follows a logistic curve and should start increasing only very slowly once you are in the 40C-50C range. If that is not the case, you might have to check that there is good thermal contact between the edge of the HL2 and the enclosure (enclosure coating removed, good thermal grease applied, downward pressure applied via the screw) as described in the wiki.

I agree that a more constant and lower temperature is better for extended part life, but we made a trade-off. At one point, I calculated the junction temperature and we were still over 20C (iirc, it is in this list somewhere) lower than the max for the 55C-60C measurements on the PCB. I think at that temperature the devices will still last a very long time. The gateware shuts off TX at 55C, so people need not be too afraid of damaging components.

73,

Steve

kf7o

[Jayson Bucknell]

Well, tonight after a few days active FT8 operation, I swapped gateware from 20200727_72p2 skimming to 20200803_72p3 and tried to start up again on 40 meters only to discover the HL output is now dead. I get indications of power output from the software and the temperature shoots up to thermal cut off really fast. (that's never happened at all before) zero output on the watt meters. I checked drive, tune.. Nothing... Receive is good to the antenna, dummy load is fine. Reverted to stable firmware. Rebooted everything.

My love affair has hit a snag. Unless it's something really botched in software I'm afraid I'm going to have to enquire about warranty or repair options. So warned.

[dick_...@hotmail.com]

Hi Steve

I found a new mechanical drawing of the 55mm case that was  previously recommended for Builds 5 through 9 up until May of 2020 when Makerfabs would supply the 40mm case.  The old "Hermes Lite 2.0 Getting Started ... " seems to have disappeared from the WIKI, but it said  "The recommended and tested enclosure for the Hermes-Lite 2 is brand SZOMK model number AK-C-C12.  A raw finish (no powder coating) is recommended for best thermal dissipation.  These enclosures are available from AliExpress, Ebay and Amazon. ...".

Attached is a copy of the Manufacturer's drawing of the AK-C-C12.dwg [it needed to be in a zip format to post].  I also made a pdf version for those who do not like to mess with dwg files.  The original is available in a zip file at: https://www.chinaenclosure.com/downfile/2015020914320507171.zip  

I believe that on some post that I read that the SZOMK AK-C-C69 was the 40mm case.  Unfortunately the best mechanicals  for this case that I could find are attached, but do not really provide adequate dimensions. Perhaps you or Makerfabs can get a dwg file like the one for the AK-C-C12  It is fairly clear from the drawings and pictures in the new WIKI page that the two cases are mechanically different in construction on the side walls.

I think this information should be in the WIKI some place, so those who would like to make modifications would have the correct dimensions to start with.  Also so users do not mistake the dimensions of the 55mm case for those used on the new 40mm case.  Might be also pertinent for end cap modifications.  When overlaying the endcap gerbers there is a difference in the hole locations.

Dick K9IVB

[Alan Hopper]

Hi Jayson,

just a thought, is the correct filter being selected? What software are you using, It might be worth trying something else just in case there is an issue there.

73 Alan M0NNB

[Jayson Bucknell]

Hi, Alan. I usually have been using SDRConsole and I hadn't changed anything since earlier in the day. I checked for anything like filters or connections / bad SWR. It's all ok. I can turn off the filters in SDRC, no change. Nothing from SparkSDR. Another Windows box also fails to transmit. I have Gnuradio flow graphs and SparkSDR in Linux and there's nothing there now either.

I see the 20200727_72p2 gateware variant disables transmit. Unless that's somehow still scrambled the transmit stages after re-flashing I'm thinking the finals have shorted.

Jayson

AA7NM

[Reid Campbell]

Hi Jayson,

Are you sure you haven't uploaded one of the Rx only variants which has 10 receivers but TX function removed?

Cheers

Reid
Gi8TME/Mi0BOT 

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/5cce787a-bbe9-4401-ba96-959b21b46e69o%40googlegroups.com.

[Alan Hopper]

Hi Jayson,

Reid beat me to it, it is worth checking what firmware the radio is actually running, most software reports this somewhere.

73 Alan M0NNB 

To unsubscribe from this group and stop receiving emails from it, send an email to herme...@googlegroups.com.

[Jayson Bucknell]

As I said at the outset, I've been through all the various gatewares. it's currently on 71.3 stable. Transmit is dead to all. Yes, I tried jiggling the wires too! hehe

Unless there's something corrupting the NVRAM from lots of flashing it's not that hard to believe the LDMOS went into runaway before/below a software threshold. The question is how does one fix it?

Jayson

AA7NM

[Alan Hopper]

Hi Jayson,

yep I just thought it was worth checking that the board was really running the gateware you think just in case something odd is going on with programming etc.  As far as I know no one has reported a LDMOS failure before and they have proven to be very robust hence exploring all avenues. What current is shown when trying to tx?

73 Alan M0NNB

[Chris Gerber]

Hi Steve

I only know the new  40mm tall enclosure and all the measuered datas where identical on two units here, also with good thermal

contact as you described, grinded of the coating.

Full 100% dutycycle power of 5.6w into a 50ohm load. It took 8min to reach 50C.

With the cooling piece added, same power conditions, it took 23min - spacing  is 5mm measured here in europe.

So there might be quit a differnce between a 55mm case versus 40mm as its a shorter way with 3,8mm to thr large bottom place.

My datas where no fake news!

Chris

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/aac409cb-d053-40f7-ae98-07dd46dab288o%40googlegroups.com.

[Chris Gerber]

Hi All 

Soory I answer again.

Here a file on which you can see what was done and tested:

 

Hermes Lite 2 Heath Dissipation

 

HL2 unit with case of 40mm from Makerfabs and 5mm gap spacing

Put together as described by the developers.

Grinded off the ripen part on the case

Made the following Output Power test and measurements:

 

Test setup:       PSU 13.80V

                        Current Meter

                        SWR Power Meter: Daiwa DP-830

                        50 Ohm 200 W Dummy load

                        Measurement monitored with PowerSDR 3.5.0 Beta 5

 

Measured Thermal Contact Surface between Board and Ripen. 35mm by 2mm (Best case)

35mm by 1mm (Worst case, if the board is not shifted to the side)

 

That is in my aspect a far to small surface to bring away the heath created with 5-6W

Expecting an efficiency of 35 to 50% in AB Mode with the mounted Final MosFets, this then would be in the order of around 6 – 9W. Quit a bit as I have seen on my few mcHF units I have built here.

 

Total Input on my Test was 13.6v and 1.47A resulting around 20W of course some power is taken of by the board etc.

Again my test shown a time of 8 min to reach 50c Temperature

 

Added my self developed Cooler Extender I could easy reach 25 min under the same test conditions as above described. That is a excellent range for all Modes used.

 

I think testing in Mode WSPR is ok, but why not using RTTY. I have seen in my long past ham life to many Radio’s produced, claiming a high output power rate and then in small letter asking not to go above 35% when on RTTY. This was very common and still is.

There is nothing said against this fabulous design and make, but it is always a way to make it good or better, that especially when no redesign is in valve.

I am very please with my HL2, also with my small changes made.

 

Regards Chris HB9BDM

 

 

Hi Steve

I only know the new  40mm tall enclosure and all the measuered datas where identical on two units here, also with good thermal

contact as you described, grinded of the coating.

Full 100% dutycycle power of 5.6w into a 50ohm load. It took 8min to reach 50C.

With the cooling piece added, same power conditions, it took 23min - spacing  is 5mm measured here in europe.

So there might be quit a differnce between a 55mm case versus 40mm as its a shorter way with 3,8mm to thr large bottom place.

My datas where no fake news!

Chris

Am 17.08.2020 um 05:43 schrieb Steve Haynal:

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/aac409cb-d053-40f7-ae98-07dd46dab288o%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/a8f3f8a4-925a-d10a-a87e-a57737b0ed0b%40swissonline.ch.

[Doug Schultz]

On my unit, received a week ago, the gap is 5mm. While some may believe the rail method is adequate, I am quite skeptical, based upon my 30 years of engineering experience. I think getting the heat out that way is a bit variable, depending on board compression against the rail, how much thermal compound is there ...

Anyway, improvements are good. I would love for someone with mechanical capabilities to fabricate some of the heat spreaders, it is an improvement for sure.

Doug

[Jayson Bucknell]

Alan, in SparkSDR there is only .1-.2 milliamps indicated on transmit. There is no change in temperature. In SDRC there is no current indication but transmit does cause a rapid rise in temperature until thermal protection cuts it off. It then cools down enough to turn on again, then off. It just bounces off the cut off at that point. Sad to think I may be the first to find this.

[Jayson Bucknell]

OK. I found the IPA current meter in SDRC. It actually shows 1750ma on transmit.

[Jayson Bucknell]

And if anyone is curious, this is a Build 9 board. Not even a month old.

[Steve Haynal]

Hi Doug and Chris,

I agree that thermal dissipation can almost always be improved. I am waiting for the day when someone will connect a PC water cooled system to the HL2 for who knows how much extra money. My point is that there is an engineering trade-off here between cost, component lifetime and component protection. A major goal of the HL2 is low cost based on a commodity ADC/DAC, and I believe the current recommended solution is a good compromise. 8 minutes key down below 50C is good for most uses. Besides the rail dissipation, there is also PCB and air dissipation that must be taken into account for any calculation. I need to go back in this list and dig up the past discussions, but I read several papers at the time and convinced myself that junction temperatures in the LDMOS were always well below the maximum rating. It is true that people can see variation, but I think adjustments can be made to be on the good side of this variation by maximizing the rail overlap, experimenting with thermal greases and ensuring there are no finishes on the enclosure where heat is intended to be transferred.

I still plan to try out Makerfabs' fabrication services, but haven't had the time to review and submit the drawings yet.

73,

Steve

kf7o

[Steve Haynal]

Hi Jayson,

Sorry to hear that your HL2 is no longer transmitting. Over 400 HL2s have been sold and this is the second (possible) instance of a LDMOS device going bad that I can recall. In the first case, the LDMOS just failed for no apparent reason soon after purchase. How deep do you want to go into the debug of this issue? Do you have an oscilloscope, current limited power supply, SMD soldering capabilities and other test equipment? I can provide a list of checks if you are interested. You also live just a few hundred miles from me. You can send the unit to me for repair if you like. I'm a bit curious why you were always seeing high temperatures. Perhaps the PA voltage regulator is bad or set wrong on your board. The PA is not run at the external voltage but is regulated to 8V. This lower voltage actually reduces the amount of heat generated.

73,

Steve

kf7o

[Jayson Bucknell]

I have some tools though possibly not enough to properly troubleshoot. Oscilloscope yes depending on the upper frequency, a current limited supply I can come by, my SMD soldering capabilities are limited. As the board itself is part of the heat spreader around the finals it's somewhat resistant to any heat I can apply. I don't have a reflow oven.

I'm definitely interesting in debugging but it if goes too deep I'll likely have to send it to you.

One thing I would like to check is is there output possible from RF1 or can one otherwise bypass the power amplifier? I'm not sure how to switch it.

73
Jayson
AA7NM

[Alan Hopper]

Hi Jayson,

you enable rf1 just by not having the pa turned on.

73 Alan M0NNB

[Jayson Bucknell]

Thanks Alan. Good to know! There's something else going on here. I noticed the PA switch in GRC and tried turning it off. Output started working. I then turned the PA back on and suddenly the adjacent receiver was picking up a very strong signal. Over in SDRC it started showing 3 watts power output with a 3-1 SWR. On a dummy load? Rewired it to the simplest possible setup. Radio>Meter>Load. It works! Meter shows 3 watts output. Good SWR, and a normal temperature swing.

It could be there's something going on with a connection in the wiring even though it looked and sounded ok at first. Maybe an amp or something between. (they're off) Sorry to make so much noise everyone. I'm going to do some more troubleshooting and try to pin it down.

I did flash between the 72.2 gateware variant overnight and 72.3 this morning. I'm very happy the finals seem to be OK. I've been really enjoying this radio the last month as was heartbroken to think I'd killed it. There's hope!

73,

Jayson

AA7NM

[Jayson Bucknell]

An oddity. I still can't get transmit working in SparkSDR on any computer. It indicates transmit and the breakout box relay clicks, but there's no output and like .1 ma current flickering on the meter. I seem to recall it working. SDRC and GRC are working fine now. So questions remain. Software?

73,

Jayson

AA7NM

[Alan Hopper]

Hi Jayson,

In Spark you need to enable the pa in the radio settings (...) button above the freqency display and have drive set above 0, can't think of anything else that would stop tx at the moment. 

73 Alan M0NNB

[Jayson Bucknell]

 OK, found it. It works. Thanks. Shutting up now.

73,

Jayson

AA7NM

[Steve Haynal]

Hi Jayson,

I'm happy to hear your radio is doing better. Maybe there is some intermittent issue such as a bad solder joint or part such as the TR relay beginning to fail. Keep us updated!

73,

Steve

kf7o

[Jayson Bucknell]

Hi Steve. I've found it. A new amplifier failed in such a way as it would pass receive signal but not transmit even in bypass or off. Weird. Remove it and the problem went away.

73,

Jayson

AA7NM

[Steve Haynal]

Hi Samuel, Chris and Group,

Thanks for the corrected drawing. Makerfabs can have these made in raw aluminum (no surface finish or treatment) for $1.80 a piece if we order 100. Once you add shipping and handling, it will probably be between $5 and $10 total price per order, shipped by Makerfabs. I do not want to distribute and ship any as it takes quite a bit of time and expense, especially for international orders which can not be shipped in a flat envelope. I am willing to use Hermes-Lite 2.0 funds to pay for an initial run. This would be for the 5mm thick heat spreader which will only work in the 40mm tall case. Please provide feedback to the following questions:

** Would you be interested in purchasing a heat spreader from Makerfabs shipped for $5 to $10?

** Do we have enough confidence in the drawing to go ahead with 100 pieces, or should we go for a smaller and probably more expensive run at first? The drawing is 5mm by 28mm by 8mm.

73,

Steve

kf7o

On Saturday, August 15, 2020 at 1:25:02 AM UTC-7, samuel kallmeyer wrote:

In case the corrected design from Chris would fit, I updated the files accordingly.

Find them attached.

Samuel, f8acb

[Ronald Nicholson]

Yes.  I would probably buy at least 2 from Makerfabs, and pay for shipping.  

The heat spreader seems to be a reasonable insurance price to pay for a decrease in the probability of component failures, etc.

Thanks, Ron n6ywu

------

[samuel kallmeyer]

Hi Steve and Group

Else if at the beginning I'm not 100% interested in purchasing  a heat spreader from Makerfabs as I'm more using the HL2 in reception (and there already was additional DHL costs with shipping HL2 in EU :( ), I will for sure contribute to this and you can add me to the list.

Regarding the drawing I provided, it would be good FMO if someone could double check the step / stl file (maybe you already did it); and when considering that the file is accurate, the risk is that the heating block dimensions are a bit lower or higher than what is provided.

If I missed something in the design, I will update it accordingly of course...

If the dimensions are lower than expected during manufacturing, people will eventually have to drill again the hole and it will dissipate less temperature then expected (without currently defining which limits)

If the dimensions are a bit higher than expected, people would have to file a bit the heater, within probably a thin tolerance.

For $5 make sense for this small aluminium part: for $10 a DIY solution can be done from a HAM :-)

73,

    Samuel f8acb

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/f1f840a8-c080-4bec-96ec-4547df00aef8o%40googlegroups.com.

[jpwa...@gmail.com]

I would be in for a couple, and if I needed to touch them up to fit with a file.

..jpw J P Watters

KC9KKO

Morris, IL USA

[Jim Ancona]

I'll commit to ordering one.

Jim 

N1ADJ

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com.

[Lou Scalpati]

I'll take 2

73

Lou KI5FTY

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAKYY9AK4s5Xs_9ejK42w3NNSJ_cz7mY_-_h7sXyJDQjfY_%3Djrg%40mail.gmail.com.

[Steve Haynal]

Hi Samuel and Group,

Thanks again for the drawing. I did review the .step file. It would be great if more people review this file. There are free .step file viewers entirely online. Makerfabs will use the .step file so that is the one to review. Here is the post with the latest .zip file:

https://groups.google.com/d/msg/hermes-lite/ePG9rzab3BM/8b_s5-U8BgAJ

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[Steve Haynal]

Hi All,

There appears to be just under 10 interested, all in the US. I may then check on ordering just 10 units as a trial and ship only to a handful of people in the US.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[Josh Logan]

I did 3D print one from the .stl file.  It's too tall to fit under the PCB in the 55mm case.  If anyones wants me to pull dimensions let me know.

73, KD7HGL

Josh

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAKYY9AK4s5Xs_9ejK42w3NNSJ_cz7mY_-_h7sXyJDQjfY_%3Djrg%40mail.gmail.com.

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/5b3cfb66-7529-42ed-9fbe-a860fcdffb94o%40googlegroups.com.

[vk4...@gmail.com]

I would be interested in an exact drawing for 55mm case. I'll get it made locally. Summer is coming soon with the 35-40 deg ambient.

73 Brian.

[alexande...@gmail.com]

I would order one directly at makerfabs with delivery to EU.

73 de

Alex, DK4FT 

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

[Chris Gerber]

HI Steve

One thing what should be observed when producing the cooler is:

The whole should be at the correct place and the diameter kept accurate at 3.5mm

Also the thicknes is rather important withe 5mm. I found a little small like 4,8 to 4.9mm is

better then to thick as with 5mm its flush with the rail, so the railpart is still used as

cooler. That gives the best result

73 Chris HB9BDM

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/71795937-12a0-4f9a-a903-35c3b324ec77o%40googlegroups.com.

[Ronald Nicholson]

In addition to heat sinking the bottom of the PCB, does anyone know if there is some sort of non-electrically-conductive but thermally conductive material can be pressed directly on top of the LDMOS devices?  Perhaps a thin thermal foam pad (as used under GPU heat sinks) or mica insulator material under a larger metal heat sink? The heat sink would have to clear the binocular transformer and relay, but otherwise could be quite large or tall, and finned.

73,

Ron

n6ywu

------

On Sunday, August 9, 2020 at 9:07:16 PM UTC-7, Steve Haynal wrote:

Hi Ron and Group,

Some people do heat sink the HL2 in other ways. There is a large exposed area on the bottom of the PCB seen in the picture below. The game is to pipe heat from that area off of the HL2. The HL2 does have temperature monitoring which will disable TX if the temperature is >55C, so you should be safe with experimenting. I like to see the HL2 at below 45C on most days, and below 50C on the hottest days.

I would like to see a small aluminum rectangle manufactured that is about the width and length of the exposed area in the picture below, with hole for the screw, and thickness that matches the distance between the bottom of PCB and enclosure surface when the HL2 is mounted in the recommended enclosure. Such a heat shim could be used in the standard enclosure to provide better thermal transfer, or for those who want to use other finned heat sinks as it would provide a spacer so that the bottom of the HL2 PCB is not shorted by the heat sink. Makerfabs does do custom CNC work, and I have HL2 credit with Makerfabs. If someone is willing to draw up such a heat shim in one of the formats accepted by Makerfabs, I can have a bunch made. A free shim goes to the person who draws it up properly.

https://www.makerfabs.com/cnc-service.html

73,

Steve

kf7o

 

On Sunday, August 9, 2020 at 11:49:41 AM UTC-7, Ronald Nicholson wrote:

If one were to operate a HL2 at 5W outside of the recommended 40mm or 55mm cases (for display purposes, scope probing both sides of the PCB, or ultra-lite SOTA/portable ops), what might be the recommended heat sinking?  

Has anyone though about or tried a plastic or plexiglass enclosure? (with some sort of heat sink).

Thanks,

Ron

n6ywu

[Steve Haynal]

Hi Josh,

Unfortunately the 55mm case clearance is different than the 40mm case. There was some discussion about this earlier in this thread. This file is only supposed to work with the 40mm case. It would be nice to have .stl or .step files for both cases.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAKYY9AK4s5Xs_9ejK42w3NNSJ_cz7mY_-_h7sXyJDQjfY_%3Djrg%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[Lou Scalpati]

I said I would order 2 but I have 2 HL2's in 55mm cases. 

73

Lou KI5FTY

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAKYY9AK4s5Xs_9ejK42w3NNSJ_cz7mY_-_h7sXyJDQjfY_%3Djrg%40mail.gmail.com.

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/5b3cfb66-7529-42ed-9fbe-a860fcdffb94o%40googlegroups.com.

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/d58c3b0d-8fae-4b3a-aa8e-bd2dec779617o%40googlegroups.com.

[Jim Ancona]

I also said I would take one, but I have a 55mm case too.

Jim

N1ADJ

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CADON%3D-5fDDSyBRf53%3DfWHkEOcnso%3D07vP7sdqrO-8Siw7_gBgA%40mail.gmail.com.

[dick_...@hotmail.com]

If you look at my post from August 12 you will find an accurate drawing for the 55mm case

https://groups.google.com/g/hermes-lite/c/ePG9rzab3BM/m/SLSoGWjSBgAJ

I am not certain that there is adequate room in the 55mm case for this design.

Would be nice if someone could get or find a similar drawing for the 40mm case.

Dick K9IVB

[Winston]

I'll try one as well.

Winston

w5ins

[Mark Wild]

I’ll take a couple for the 40mm case.

Mark G6DDX

[Sid Boyce]

Hi Ron,
I use thick copper plate and on my original HL2, 2 layers of thin copper
plate that bears down on the LDMOS devices and bolted to the side of the
case.
This arrangement was the original recommended by Steve and I installed
on all 3 of my HL2's.
73 ... Sid.

On 26/08/2020 17:40, Ronald Nicholson wrote:
> In addition to heat sinking the bottom of the PCB, does anyone know if
> there is some sort of non-electrically-conductive but thermally
> conductive material can be pressed directly on top of the LDMOS

> devices?  Perhaps a thin thermal foam pad (as used under GPU heat

> hl2bottom.jpg
>
>
>
> Â

>
> On Sunday, August 9, 2020 at 11:49:41 AM UTC-7, Ronald Nicholson
> wrote:
>
> If one were to operate a HL2 at 5W outside of the recommended
> 40mm or 55mm cases (for display purposes, scope probing both
> sides of the PCB, or ultra-lite SOTA/portable ops), what might

> be the recommended heat sinking? Â

>
> Has anyone though about or tried a plastic or plexiglass
> enclosure? (with some sort of heat sink).
>
> Thanks,
>
> Ron
> n6ywu
>

> --
> You received this message because you are subscribed to the Google
> Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to hermes-lite...@googlegroups.com

> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hermes-lite/784c0421-8b53-43f9-b95b-16ee79e729dbo%40googlegroups.com
> <https://groups.google.com/d/msgid/hermes-lite/784c0421-8b53-43f9-b95b-16ee79e729dbo%40googlegroups.com?utm_medium=email&utm_source=footer>.


--
Sid Boyce ... Hamradio License G3VBV, Licensed Private Pilot
Emeritus IBM/Amdahl Mainframes and Sun/Fujitsu Servers Tech Support
Senior Staff Specialist, Cricket Coach
Microsoft Windows Free Zone - Linux used for all Computing Tasks

[Steve Haynal]

Hi Group,

Makerfabs has agreed to make 10 for evaluation. I will ship a few to US domestic testers only. Again, this is only for the 40mm enclosure that Makerfabs sells. Makerfabs agreed to include a heat spreader in future enclosure sales. They may be able to sell them one off too. We are nearing the end of the last batch of 100 units and preparing for the next batch. 

For possible 55mm tall enclosure support, I need a drawing in the format accepted by Makerfabs (.stl .step).

73,

Steve

kf7o

[Jaroslav Škarvada]

Hi,

if you by chance prototype the 55 mm version in the future (with
Makerfab), I will be interested

73! Jaroslav, OK2JRQ

Steve Haynal napsal(a):

> Hi Josh,
>
> Unfortunately the 55mm case clearance is different than the 40mm case.
> There was some discussion about this earlier in this thread. This file
> is only supposed to work with the 40mm case. It would be nice to have
> .stl or .step files for both cases.
>
> 73,
>
> Steve
> kf7o
>
>
>
> On Tuesday, August 25, 2020 at 10:27:43 PM UTC-7, Josh Logan wrote:
>
>
> I did 3D print one from the .stl file.  It's too tall to fit under
> the PCB in the 55mm case.  If anyones wants me to pull dimensions
> let me know.
>
> 73, KD7HGL
> Josh
>
>
>
>

> hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit
> https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com

> <https://groups.google.com/d/msgid/hermes-lite/110e6a90-f742-4d78-b7ff-51c67946de5dn%40googlegroups.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to
> the Google Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails
> from it, send an email to

> hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hermes-lite/CAKYY9AK4s5Xs_9ejK42w3NNSJ_cz7mY_-_h7sXyJDQjfY_%3Djrg%40mail.gmail.com
> <https://groups.google.com/d/msgid/hermes-lite/CAKYY9AK4s5Xs_9ejK42w3NNSJ_cz7mY_-_h7sXyJDQjfY_%3Djrg%40mail.gmail.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the
> Google Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails from

> it, send an email to hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hermes-lite/5b3cfb66-7529-42ed-9fbe-a860fcdffb94o%40googlegroups.com
> <https://groups.google.com/d/msgid/hermes-lite/5b3cfb66-7529-42ed-9fbe-a860fcdffb94o%40googlegroups.com?utm_medium=email&utm_source=footer>.

>
> --
> You received this message because you are subscribed to the Google
> Groups "Hermes-Lite" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to hermes-lite...@googlegroups.com
> <mailto:hermes-lite...@googlegroups.com>.

> To view this discussion on the web visit

> https://groups.google.com/d/msgid/hermes-lite/d58c3b0d-8fae-4b3a-aa8e-bd2dec779617o%40googlegroups.com
> <https://groups.google.com/d/msgid/hermes-lite/d58c3b0d-8fae-4b3a-aa8e-bd2dec779617o%40googlegroups.com?utm_medium=email&utm_source=footer>.

[Steve Haynal]

Hi Group,

See the picture of the "heat shims" from Makerfabs. I tried one out and I do see some improvement. It took 11 minutes to reach 50C during 5W WSPR transmits. The enclosure heated up nicely so I think there is good thermal transfer. The hole didn't quite line up so I had to drill the hole slightly larger. I sanded the bottom enclosure finish off where it touches the heat shim. I used thermal grease on both contacts. I'll send a few out to domestic testers.

73,

Steve

kf7o

[Samir, OD5SK]

Hello Steve,

I got my HL2 yesterday, I am so happy with the performance but of course the issue I had is getting the HL2 temperature as low as possible as the protection circuit keeps shutting down the tx and  I don't want to risk burning the finals.  Regarding the " Heat Shims " , are they going to be available soon for purchase ? On the other hand I am thinking of adding a fan ( 8 x 8 cm ) , apart the small fan connection method, is there a connection on the board for such a fan.

Thank you and Best 73,

Samir

OD5SK , HZ1SK

[Steve Haynal]

Hi Samir,

I am glad your are enjoying your HL2. Have you followed the assembly procedure on the wiki?

https://github.com/softerhardware/Hermes-Lite2/wiki/Final-Assembly#assembly-procedure

This provides adequate heat dissipation for most uses.

See the end plate pages linked to on the IO wiki page for fan circuit information:

https://github.com/softerhardware/Hermes-Lite2/wiki/IO#io-boards

A few people are testing the heat shim. I'm not sure if or when it will be available from Makerfabs. Makerfabs is considering changing the enclosure and may add something to a new enclosure.  

73,

Steve

kf7o

[Samir, OD5SK]

Hi Steve,

Thank you for your response and I really appreciate it . Yes indeed, I followed the assembly procedure but thought its better if I can bring the temperature down a bit further as we live in a hot area here. I wrote a small review on the Hermes Lite 2 in Arabic at :

http://www.laselki.net/vb/showthread.php?t=12245

Best 73s,

Samir

OD5SK , HZ1SK

[Cesc Gudayol]

Hi guys

Related with the fan, I have a newbie question: I'm assembling only the Fan Control part of the back endplate, and I want to double-check that the only components I need are:

– R1: 10k

– R18: 1K

– C4: 100nF 

– C5: 100nF

– Q1: DTC144E

– Q2: IRLM6402 

I understand that the only connections that I must do from the HL2 board are as follows:

(I think I don't need the Vlvds connection for the Fan circuit)

Am I right? Am I missing something?

Thanks for your help.

73, Cesc

EA3IGT

[ron.ni...@gmail.com]

I ran a small experiment to see how well heat transfer from the Hermes Lite 2 PCB to the side of the 40mm case was helping lower the working temperature.

First I used the HL2 for receive-only for 4 minutes: temp =  30.5 C =  87.0 F

Then I transmitted a wspr signal at 30 dBm (1 Watt).

Initial temperature rise after 2 minutes was: temp =  37.6 C =  99.7 F

After 10 minutes: temp =  42.1 C = 107.7 F

After 1 hour: temp =  47.1 C = 116.8 F

Then I immediately placed one of those small battery-operated personal fans about 4 cm away from the right side of the exterior of the 40mm case, and set the fan on "low" (while still continuously transmitting 20M wspr at 30dBm).

After 2 minutes: temp =  40.1 C = 104.2 F

After 10  minutes: temp =  34.2 C =  93.6 F

That's a 13 C temperature drop from peak, just by adding a bit of airflow around the case exterior.  So the thermal conduction from the PCB slot to the case exterior seems to be significant. I wonder if just bolting a large (passive noiseless) external heat sink to the right side of the HL2 case would significantly increase thermal dissipation.

73,

Ron

n6ywu

On Sunday, August 9, 2020 at 9:07:16 PM UTC-7 softerh...@gmail.com wrote:

Hi Ron and Group,

Some people do heat sink the HL2 in other ways. There is a large exposed area on the bottom of the PCB seen in the picture below. The game is to pipe heat from that area off of the HL2. The HL2 does have temperature monitoring which will disable TX if the temperature is >55C, so you should be safe with experimenting. I like to see the HL2 at below 45C on most days, and below 50C on the hottest days.

I would like to see a small aluminum rectangle manufactured that is about the width and length of the exposed area in the picture below, with hole for the screw, and thickness that matches the distance between the bottom of PCB and enclosure surface when the HL2 is mounted in the recommended enclosure. Such a heat shim could be used in the standard enclosure to provide better thermal transfer, or for those who want to use other finned heat sinks as it would provide a spacer so that the bottom of the HL2 PCB is not shorted by the heat sink. Makerfabs does do custom CNC work, and I have HL2 credit with Makerfabs. If someone is willing to draw up such a heat shim in one of the formats accepted by Makerfabs, I can have a bunch made. A free shim goes to the person who draws it up properly.

https://www.makerfabs.com/cnc-service.html

73,

Steve

kf7o

On Sunday, August 9, 2020 at 11:49:41 AM UTC-7, Ronald Nicholson wrote:

If one were to operate a HL2 at 5W outside of the recommended 40mm or 55mm cases (for display purposes, scope probing both sides of the PCB, or ultra-lite SOTA/portable ops), what might be the recommended heat sinking?  

[Steve Haynal]

Hi Cesc,

Yes, those are the only components required for the fan. Vlvds is not need for the fan. You will have to connect ground also. You can also take Vsup and GND from the back IO strip of the HL2 instead of DB1 if it is easier.

73,

Steve

kf7o

[Steve Haynal]

Hi Ron,

It is interesting to hear that the small fan worked well for you. I've tried a very small internal fan for the 55mm case but didn't notice as much improvement as you. I suspect your fan offers more air flow.

73,

Steve

kf7o

[Steve Haynal]

Hi Samir,

Is the protection circuit shutting down the TX for you when reaching 55C? What is your ambient temperature? It would be interesting to know what your starting unit and ambient temperature is, and then how long it takes for the unit to reach 50C if you transmit a full 5W into a dummy load. We can then compare your results with others.

73,

Steve

kf7o

[ron.ni...@gmail.com]

Hi Steve,

Yes. It's very possible that a small fan outside the case has less obstructed airflow, plus a larger cooling surface consisting of portions of the metal case wall both above and below the case slot.  The screw presses the PCB against the bottom of the slot, so cooling the exterior wall of the case below the slot probably helps remove additional heat.  

Thus, I wouldn't be surprised if one of those mini heat sink + fan combos taped to the side of the case would provide as much or more cooling than a fan inside the case.  It would be easy to tape two of the small ones made for Raspberry Pi's to the side of the case.

73,

Ron

n6ywu

[Winston]

Regarding the heat shims:  I attached a 0.2mm layer of thermal adhesive between the shim and bottom of the PCB to increase heat transfer efficiency.  The thickness of the adhesive is not a problem but it appears the shim is a bit too wide because it is up against two PCB components that could cause a short.  I would like to have also applied adhesive to the bottom and side of the shim but since the PCB slides into the case grooves, the friction would probably stop the PCB from sliding in fully.

[Chris Gerber]

Hi

As HL2 newbie I see discussion about using Fan etc. with parts on the backplate

But I never seen a real drawing, scematic of it, also for the use of the printed backplate

for the 40mm enclosure?

73 Chris

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/00414ee9-c68f-484d-9955-4e83bfbf31cco%40googlegroups.com.

[Samir, OD5SK]

Hi Steve,

I have added a thermal compound to the joint and now things are much better, temperature is reaching 45 C after 9 minutes.

Best 73,

Samir

OD5SK , HZ1SK

[keyboa...@gmail.com]

Hi all. I received my shim in the mail today and installed it. It seems to clear the other components ok. It was a very tight fit but I managed to slide it into place along with the board and some white silicone heatsink compound on the underside of the board and the on the inside of the case after I removed the anodization with sand paper.

It makes a huge difference! 5 watts into a dummy load for over 30 minutes no problem. The maximum temperature was a slowing climb to 48 degrees C after which it leveled off. Pointing a fan at the case and blowing air across it caused the temperature to fall back down to 40 degrees and was still slowly dropping.

That takes care of the one real usage issue I was having. I imagine other shim versions will come along or a possible case redesign will solve the problem more generally.

Thanks to all who helped out! Hermes-Lite Rocks!

73,
Jayson
AA7NM

[Steve Haynal]

Hi All,

Just a few comments. 

Samir, glad the thermal compound helps. I've actually noticed differences between thermal compounds - some work better than others. It would be great if people could report which brand(s) work well.

Chris, you can find a links to all the IO information including gerber files and schematics from the wiki page:

https://github.com/softerhardware/Hermes-Lite2/wiki/IO#io-boards

Winston and Jayson, glad your heat shims arrived and that Jayson's works well. I had to drill the hole in mine slightly larger for it to fit in the case as it could not be pushed close enough to the outside edge. . It was close to other components. I used thermal paste so very thin interface. I'd appreciate feedback from testers on any changes to make it fit better. I will send an updated drawing to Makerfabs as they agreed to include this in future orders. Unfortunately, I don't think they will modify the enclosure.

73,

Steve

kf7o

[Probir]

One BEST way of using Thermal paste is PASTE AND REMOVE.

 

Use a thin layer of thermal paste – since all is manual certainly there will be extra thermal paste on the surface – remove the extra thickness of the paste from the surface through a thin Blade.

 

Please note thermal paste is use just to fillup the microscopic holes not visible through naked eye.

 

IF ANY ONE here wants to add more on to it – I WOULD like to learn.

 

73

Probir, VU2BQF

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/65c0d209-333e-48e8-a328-b27daf3cb66co%40googlegroups.com.

[Chris Gerber]

Thanks Steve, all checked and stored away

73 Chris HB9BDM

--
You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/65c0d209-333e-48e8-a328-b27daf3cb66co%40googlegroups.com.

[Samir, OD5SK]

Hello All,

I found on AMAZON a nice Heatsink that can be attached to HL2 case , I wonder what will be the effect on the temperature but I think it will improve cooling of the enclosure and heat dissipation. Dimension is 15 cm x 7 cm , it can be attached to the top of the box using thermal tape.

https://www.amazon.com/Aluminum-Heatsink-Amplifier-Transistor-Semiconductor/dp/B07VMMJCLY/ref=sr_1_7?dchild=1&keywords=heatsink&qid=1600279808&sr=8-7

Best 73,

Samir

OD5SK , HZ1SK

[keyboa...@gmail.com]

Hello, All. For details on the process of installing the shim I should say
the fit was EXTREMELY tight in that space. The HL board now pushes
against the top of the groove and in fact didn't fit initially as I'd
tried adding a bit of solder along the top board edge in an attempt to
improve coupling. Removed and it's fine.

I set the board in the groove enough to slide the shim between the board
and the case underneath the finals and then slid them together down to
the case hole using a toothpick though their holes. That kept from
having any pressure against components by the shim as it moved into place.

The case hole was hand measured and drilled (not currently countersunk)
and while the standard M3 screw was able to get all the way through it's
tight through the hole and I couldn't tighten it down until the end
plates were set and HL2 and the N2ADR filter board aligned in the case
properly. After that it's a solid sandwich. Thermal compound is
naturally a very thin layer.

I guess I now appreciate the hesitancy on consulting other maker
machinists to craft something like this by description alone. The
tolerances are extremely tight. You might consider shaving a few
micrometers in thickness.

The improvement in thermal equilibrium is tremendous. The duty cycle is
100%. It's slower to heat and levels off well below cutoff. Cools
quickly. A variety of external cooling could be applied to the case. Peltier Junction anyone?

Reading the Hermes wiki and development history is a bit like coming
across a great series in a later season and trying to binge back bits to
understand how things came to be the way they are now. My perspective is
jumbled but the story's gotten better over time. Keep up the great
work!

73,
Jayson
AA7NM

[Steve Haynal]

Hi Jayson,

Thanks for the feedback. I should have measured all the shims before sending them out. I suspect the larger contribution to variation is the enclosure. I'd like refine into a design that works well across all enclosure variations.

73,

Steve

kf7o

[Steve Haynal]

Hi Samir,

It would be interesting to see how that works. Are you still having overheating issues? The farther the heatsink is from the source of heat, the more metal the heat must travel through and be affected by thermal resistance. My guess is that a heatsink like this would have to be mounted as close as possible to the LDMOS devices to reduce the thermal resistance and maximize its dissipation.

73,

Steve

kf7o

[Samir Khayat]

Hi Steve,

I am not having issues inside the house but I was thinking about using the HL2 outdoor where the ambient temperature here is very high ( 43 C today ) , so this is why I guess additional ways that will help in heat dissipation are required.Yes you are right about the location of the heat sink , there is other types which are smaller in size and can be mounted on the side close to the PA transistors.

Best 73,

Samir

OD5SK , HZ1SK

You received this message because you are subscribed to a topic in the Google Groups "Hermes-Lite" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/hermes-lite/ePG9rzab3BM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/ebbe339f-97b3-4798-81db-3d25807458cbo%40googlegroups.com.

[Probir]

Hi Samir,

 

The Enclosure heat is a cumulative heat from the LDMOS and other parts inside. So there will be latency in the heat transfer.

 

For outside application it will be good if you can use a FAN at the rear panel and ensure the air flow is outside.

 

 

73

PROBIR vu2bqf

.........................................

 

From: Samir Khayat

Sent: Thursday, September 17, 2020 12:16 PM

To: Steve Haynal

Cc: Hermes-Lite

Subject: Re: Minimum heat-sinking requirements? (completely outside of the metal case)

 

To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/CAFY1RUOr-Mks1EJf%3D9Gcq%2B9Y9FtanX7ZiWT0oh5me_xiyoHHzw%40mail.gmail.com.

[rhqq2yxrkt]

Thermal compounds / heat sink compounds have a specification watts per meter Kelvin w/mk.  The higher the number the greater the heat transference or heat coupling.

John G3UGY

[Steve Haynal]

Hi John,

Yes, I see that listed on Mouser. I think my last batch has a lower w/mk than what I was using previously.

73,

Steve

kf7o

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite+unsubscribe@googlegroups.com.

[ron.ni...@gmail.com]

Hi Steve,

Before modifying the heat sink shim, I first tried modifying the case by sanding off the case's black finish under and to the side of where the heat sink would need to slide in and contact the case bottom.  That alone was almost enough. In addition I needed to de-burr just the very edges of the heat sink. Then it would then slide in with the PCB.  

Now the issue is that with an M3 screw through the PCB and heat sink hole, the heat sink chamfer is holding the board away from being completely set into the case slot by nearly 1 mm.  It looks like I may need to Dremel a groove into the heat sink chamfer to make it more of a rectangular or circular inset, rather than flat diagonal, in order to clear the bottom case rail; and thus allow the PCB to press firmly into the case slot on the power transistor side of the PCB. 

I'm currently trying to figure out some kind of jig that will allow my to cut a clean groove in the chamfer of the required depth.

73,

Ron

n6ywu

[Steve Haynal]

Hi Ron,

This sounds similar to what I ran into. I drilled a larger hole in the heat sink shim so that the M3 screw would fit.

From various input, I'm thinking the height of the shim needs to be reduced by 0.2mm, 0.1mm off the top and bottom, and the chamfer pulled in by 0.5mm.

73,

Steve

kf7o

[Winston]

I spent a good amount of time filing down the diagonal and short edge of the shim.  I would say 1mm would be a good amount to reduce to account for any tolerance of the hole and case.

Winston

w5ins

[Chris Gerber]

Hi Steve

With interest am I reading the mails. So desiged to write an instruction to show how easy it is, to install the shim.

Important is that the shim is accurately poduced. As with my given datas it will fit perfect.

I checked it out several times on different units. Drilling large holes or using thinner screws will missplace the give position

the shim. Also it should be checkes for no sharp edges along the side facing the enclosure corner. Otherwise round or file off.

This shim correct set is the easyest and cheapest way to solve the problem.

Please check the attachment...

 73 Chris HB9BDM

 

 HL2 Shim Mounting Instruction for 40mm Enclosure

 

Chris HB9BDM

 

Step 1: Place the Hermes and Filter board in correct position and drill then a

 3.5mm hole, according to former instructions.

 

Step 2: Use sandpaper to grind of the coating at the bottom of the enclosure,

also the shoulder part of the rail. This will assure good thermal contact.

Then coat the places with Thermal compound, also the Rail Shoulder and side.

 

Step3: Pic1, Positioning the enclosure accordingly.

 

                                      Pic1

 

 

Step 4: Pic2, Check thickness of the shim, it should be 5.00mm or 0.1mm less.

Coat the board surface where the shim is placed also with Thermal compound.

Place the shim between board and enclosure, so it still sticks out (Pic1)

Narrower side of shim facing board, wider side of shim facing enclosure.

Turn the enclosure together with the board, shim in-between carefully around

(Pic2) and use the 3mm mounting screw to hold shim in place during the sliding

in procedure, till you reach the mounting whole position.

 

 

 

 

 

                                            Pic2

 

Step 5: Pic3 carefully remove the mounting screw and insert form the bottom

of the enclosure. Secure with washer and nut, but don’t tighten.

 

                                      Pic3

 

Step 6: Connect Hermes and Filter Boards. Then press the whole assembly fully onto

the side, to ensure a good thermal contact with the side of the enclosure. Then tighten

the 3mm screw. That concludes the job.

[ron.ni...@gmail.com]

Hi Chris,

You may be putting the heat sink shim in upside down compared to the orientation that others are trying.

My assumption is that the goal is to get the wider part of the shim closer to the edge of the board where all the thru-hole vias are conducting heat away from the power transistors.  Unfortunately, the chamfer needs to be cut about 1 mm deeper to make that happen.

73,

Ron

n6ywu

[ron.ni...@gmail.com]

Hi Steve,

I did not find the shim needed to be reduced in height at all.  Just some de-burring the edges of the shim, and sanding some paint off the bottom of the case, allowed the shim to slide in easily with the PCB.  Removing some paint from the inside bottom of the case seems like it would improved heat transfer (thru some thermal compound), similar to rationale for sanding off the paint in the slot.

But the current chamfer does interfere with the slide rail, and thus doesn't allow the screw to be inserted, with the PCB all the way into the side rail, without either cutting an approx. 1mm V groove into the chamfer, or grinding out a portion of the side rail near the screw location.  I may try the latter to maximize the volume of the shim.

73,

Ron

n6ywu

[Chris Gerber]

Hi Ron

Seems you know every thing better. I have designed the shim for me. Tested at 5 HL2, 40mm units

and they worked perfect. With up to 30min 5W full power into a dummy load to reach a temp.

of 50C. Placed exactly as shown in the description. Easy to fabricate and install. But it does need some

accuracy to drill the hole and place the shim, here that was possible.

But seems you know better, so fine, why did you not have any idea right away?

The HL2 is a fime unit and did not need many correction, but....

This is end of line for any help from me.

Gl es 73 Chris

So you do what you like for me it theend of any line

--

You received this message because you are subscribed to the Google Groups "Hermes-Lite" group.

To unsubscribe from this group and stop receiving emails from it, send an email to hermes-lite...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hermes-lite/816b5fa7-1a70-4662-8c2e-00bb99370a55n%40googlegroups.com.