Have E* send you a safe one. You have a fire hazard or at best an electronic component will fry killing it totally.seandudley said:
Average 622/722 Temperature thread, Is your's Hot?
- Thread starter cincyguy2k4
- Start date
- Latest activity Latest activity:
- Replies 1K
- Views 196K
power tap
Mr. Aroid
Nice job on the the fan mod. Where did you pull power for the new fan?Mark_AR said:
Mr. Aroid
622 Temp. Measurements and Mitigation Trials
Hello, all! Well I finally got around to doing some 622 temperature testing and mitigation trials of my own, like Mark_AR has been doing. I wanted a better, "real-time" feedback so I did some temp. probing vs. waiting for the results in the HDD temp. readings counters. I won't comment again on my motivation for this effort - I think most of us agree on that!
DISCLAIMER: I took some personal and equipment risks doing this! Don't attempt doing anything inside a unit unless you have a good idea of what you're doing and the associated risks, and unless you are willing to accept the consequences of your actions. Opening a unit may void the warranty. (I did not break the internal warranty seal covering a fan bracket mounting screw nor do any disassembly other than removing the cover.)
This is my replacement "E" version unit. In a nonscientific assessment it does seem to run a bit cooler to the touch on the outside vs. my returned "D" version. Upon initial internal inspection there does not seem to be any visible difference between my unit and Mark's (not sure what version he has). There are no additional heatsinks, vents, or fans or such.
I used a Sears 34.82400 autoranging DMM with the K-type thermocouple that comes with it (see photos). The 622 was powered at all times - I was actually watching a program via HDMI throughout these trials (immediate feedback for any "whoops" events!) but not recording anything other than what the HDD is doing for normal viewing. My counters initially read 125 HI, 84 LO, 104 AVG. They were the same after these trials. The ambient was around 74F when I started. (It actually rose to over 80F, but the wall thermometer is close to the TV that had been off prior to the first reading.) At no time was the 622's internal fan seen to be running. I never saw any unusual changes in the HD PQ, SS audio, or performance that might be attributable to my "invasion".
I quickly realized that there are MANY heat generators in this unit, not just the HDD! There are 3 power semiconductors with heatsinks in the power supply that were all too hot to touch, and another on the main PCB that I didn't try. And ALL of the Broadcom chips were VERY hot, particularly the large one just below/behind the HDD. That one seemed the hottest so I focused on it for these initial trials.
I checked the temp. probe (poor-man's calibration!) by first inserting it into the ice on the freezer right next to the 622 (~30F) and then into my hot tub (~95F, matched the thermometer +1deg.) and those readings seemed in line.
I first measured the temp. on the upper rear case of the HDD - about 118F. I then measured the BC chip. The TC tip was just touching the top of the BC chip on its "silver" part in about the center (photo1). At that time (with the cover off about 15 minutes) the chip was about 130F, almost too hot to touch. After some lack of repeatability in my measurements I realized that where I measure on that chip made a difference, so I had to repeat my measurements. The results below were based on a relative "hot spot" on the chip and I did my best to make sure the contact point and "pressure" remained constant.
I wanted to test the unit with the cover on, off, and with some added cooling. The cooling was primarily accomplished with the cover off and a very cheap 3-fan laptop cooler blowing down into the unit (photo 2). This is a very mild airflow (I forget the volume, but I do have the spec. sheet), probably a good bit less than what the internal fan can provide. After several trials I obtained reasonably repeatable results (+/- 3-5F) as follows:
Cover on: ~160F
Cover off: 130-135F
Fans over open unit, off: ~150F
Fans over open unit, on: 100-105F
Cover on - one fan blowing in from right side vents: ~155F
This morning after 6+hrs standby, fans running (ambient 68F): 99F
Some reasonable observations/conclusions/commentary:
1) The BC chips are major contributors to the internal heat. (At one time the meter indicated 170F (!) on that chip, but I couldn't repeat that measurement.)
2) Removing the cover alone allows the tested chip to cool by about 25F.
3) Air moving over the chip(s) is indeed necessary (as Mark_AR reported) to obtain significant cooling. In my experiment that amounted to almost 60F !! (Part of that may be due to the fact that the top was still partially open.) I also believe that flow needs to be ambient air from outside the case, not just internal air moved around, altho' that alone probably helps somewhat.
4) The internal fan is doing essentially NOTHING to help the situation. I believe it may operate only during extreme situations, but how hot does it need to be? Close to spontaneous combustion, as one poster feared ?? (There seems to be considerable variabillity here as ralfyguy reported that his fan is always running. How can we manage that in all these units? No apparent adjustment.)
5) Apparently modest swings downward in ambient temp. / standby time don't make much difference in the low temp. readings.
Some additional questions / ideas for further investigation:
1) How much heat does the HDD contribute? Does its temp. vary measurably with additional activity, i.e., if it's "very busy" with multiple R/Ws due to recording multiple programs, etc. will its temp. increase considerably?
2) What would heatsinking the BC chips do to lower their temps.? (I don't think that alone will help much - there also needs to be moving air.)
3) How much will the large BC chip's temp. change if it is "busy" with MPEG-4 decoding or such. (I'm assuming that's where the decoding activity is.)
Where I will go from here:
I plan to make an out-board fan system using 3 or 4 of those fans shown mounted into a block of styrofoam or such and held in place to cover the vents on the left (power supply) side. I can then vary the direction of flow and the number of fans running to see what makes the most difference in the internal readings. I might also place a single fan over the right side vents to work in conjunction with or opposing the left-side fans. I'm also considering another internal fan similar to what Mark did, but perhaps focused to the hottest chip.
I'll leave the TC inside the unit for now (place and mounting method TBD, but probably held more firmly on the top of that same BC chip) so I can take frequent readings. I am concerned about providing adequate filtering for the outside air, which then also leads to a question about what happens if the filters get clogged-up? Probably worse than no fans!
I hope this provides some additional insight and motivation for further experimentation to find the most viable mitigation techniques to solve our universal "problem". At least it's a great discussion/debate!
More to follow. Thanks for reading and BRgds...
Hello, all! Well I finally got around to doing some 622 temperature testing and mitigation trials of my own, like Mark_AR has been doing. I wanted a better, "real-time" feedback so I did some temp. probing vs. waiting for the results in the HDD temp. readings counters. I won't comment again on my motivation for this effort - I think most of us agree on that!
DISCLAIMER: I took some personal and equipment risks doing this! Don't attempt doing anything inside a unit unless you have a good idea of what you're doing and the associated risks, and unless you are willing to accept the consequences of your actions. Opening a unit may void the warranty. (I did not break the internal warranty seal covering a fan bracket mounting screw nor do any disassembly other than removing the cover.)
This is my replacement "E" version unit. In a nonscientific assessment it does seem to run a bit cooler to the touch on the outside vs. my returned "D" version. Upon initial internal inspection there does not seem to be any visible difference between my unit and Mark's (not sure what version he has). There are no additional heatsinks, vents, or fans or such.
I used a Sears 34.82400 autoranging DMM with the K-type thermocouple that comes with it (see photos). The 622 was powered at all times - I was actually watching a program via HDMI throughout these trials (immediate feedback for any "whoops" events!) but not recording anything other than what the HDD is doing for normal viewing. My counters initially read 125 HI, 84 LO, 104 AVG. They were the same after these trials. The ambient was around 74F when I started. (It actually rose to over 80F, but the wall thermometer is close to the TV that had been off prior to the first reading.) At no time was the 622's internal fan seen to be running. I never saw any unusual changes in the HD PQ, SS audio, or performance that might be attributable to my "invasion".
I quickly realized that there are MANY heat generators in this unit, not just the HDD! There are 3 power semiconductors with heatsinks in the power supply that were all too hot to touch, and another on the main PCB that I didn't try. And ALL of the Broadcom chips were VERY hot, particularly the large one just below/behind the HDD. That one seemed the hottest so I focused on it for these initial trials.
I checked the temp. probe (poor-man's calibration!) by first inserting it into the ice on the freezer right next to the 622 (~30F) and then into my hot tub (~95F, matched the thermometer +1deg.) and those readings seemed in line.
I first measured the temp. on the upper rear case of the HDD - about 118F. I then measured the BC chip. The TC tip was just touching the top of the BC chip on its "silver" part in about the center (photo1). At that time (with the cover off about 15 minutes) the chip was about 130F, almost too hot to touch. After some lack of repeatability in my measurements I realized that where I measure on that chip made a difference, so I had to repeat my measurements. The results below were based on a relative "hot spot" on the chip and I did my best to make sure the contact point and "pressure" remained constant.
I wanted to test the unit with the cover on, off, and with some added cooling. The cooling was primarily accomplished with the cover off and a very cheap 3-fan laptop cooler blowing down into the unit (photo 2). This is a very mild airflow (I forget the volume, but I do have the spec. sheet), probably a good bit less than what the internal fan can provide. After several trials I obtained reasonably repeatable results (+/- 3-5F) as follows:
Cover on: ~160F
Cover off: 130-135F
Fans over open unit, off: ~150F
Fans over open unit, on: 100-105F
Cover on - one fan blowing in from right side vents: ~155F
This morning after 6+hrs standby, fans running (ambient 68F): 99F
Some reasonable observations/conclusions/commentary:
1) The BC chips are major contributors to the internal heat. (At one time the meter indicated 170F (!) on that chip, but I couldn't repeat that measurement.)
2) Removing the cover alone allows the tested chip to cool by about 25F.
3) Air moving over the chip(s) is indeed necessary (as Mark_AR reported) to obtain significant cooling. In my experiment that amounted to almost 60F !! (Part of that may be due to the fact that the top was still partially open.) I also believe that flow needs to be ambient air from outside the case, not just internal air moved around, altho' that alone probably helps somewhat.
4) The internal fan is doing essentially NOTHING to help the situation. I believe it may operate only during extreme situations, but how hot does it need to be? Close to spontaneous combustion, as one poster feared ?? (There seems to be considerable variabillity here as ralfyguy reported that his fan is always running. How can we manage that in all these units? No apparent adjustment.)
5) Apparently modest swings downward in ambient temp. / standby time don't make much difference in the low temp. readings.
Some additional questions / ideas for further investigation:
1) How much heat does the HDD contribute? Does its temp. vary measurably with additional activity, i.e., if it's "very busy" with multiple R/Ws due to recording multiple programs, etc. will its temp. increase considerably?
2) What would heatsinking the BC chips do to lower their temps.? (I don't think that alone will help much - there also needs to be moving air.)
3) How much will the large BC chip's temp. change if it is "busy" with MPEG-4 decoding or such. (I'm assuming that's where the decoding activity is.)
Where I will go from here:
I plan to make an out-board fan system using 3 or 4 of those fans shown mounted into a block of styrofoam or such and held in place to cover the vents on the left (power supply) side. I can then vary the direction of flow and the number of fans running to see what makes the most difference in the internal readings. I might also place a single fan over the right side vents to work in conjunction with or opposing the left-side fans. I'm also considering another internal fan similar to what Mark did, but perhaps focused to the hottest chip.
I'll leave the TC inside the unit for now (place and mounting method TBD, but probably held more firmly on the top of that same BC chip) so I can take frequent readings. I am concerned about providing adequate filtering for the outside air, which then also leads to a question about what happens if the filters get clogged-up? Probably worse than no fans!
I hope this provides some additional insight and motivation for further experimentation to find the most viable mitigation techniques to solve our universal "problem". At least it's a great discussion/debate!
More to follow. Thanks for reading and BRgds...
Attachments
Last edited:
Excellent report, Bhelms! 160 F for the Broadcomm chip? That's is way too hot. I would say that is the main problem right there. A heatsink will alone will not help, like you said this unit needs air flow (massive air flow).
So can we assume that the BC chip runs 30-35F degrees higher than the hard drive?
Does the 622 have any kind of vents at the bottom? I have my 625 sitting on top of the laptop cooler and the hot air gets sucked out from the bottom of the 625. My Low temp is at 68F (for the hard drive). I know that we're talking about two different animals but the concept applies. The hot air needs to get out of the 622 somehow for it to run healthy.
So can we assume that the BC chip runs 30-35F degrees higher than the hard drive?
Does the 622 have any kind of vents at the bottom? I have my 625 sitting on top of the laptop cooler and the hot air gets sucked out from the bottom of the 625. My Low temp is at 68F (for the hard drive). I know that we're talking about two different animals but the concept applies. The hot air needs to get out of the 622 somehow for it to run healthy.
Tks, RandallA (Took me long enough!) It's tough to assume anything, but based on these preliminary measurements I have no doubt that what is being reported in the Counters is NOT very close to the temps. at least that one chip is seeing, and probably the rest of them as well! The other hottest ones (by touch, not actual measurement) are two identical chips on the PC just in front of the two sat. tuners. Where Mark_AR placed his fan will do a very good job cooling those, less so the one I focused on.
The bottom is slotted all over, but the PCB itself blocks any direct flow from them. The natural convection would probably be just up the sides of the case and out the various side and back panel openings. That won't provide any direct airflow over those components that need it most! They're sitting in essentially still air unless something forces it.
Later...
The bottom is slotted all over, but the PCB itself blocks any direct flow from them. The natural convection would probably be just up the sides of the case and out the various side and back panel openings. That won't provide any direct airflow over those components that need it most! They're sitting in essentially still air unless something forces it.
Later...
i've been suspecting the Broadcomm chip being too hot is the cause for video stuttering/HDMI issues, after all isn't that the chip's job, decoding video? I wish dish would enable that internal fan to turn on at the lowest speed possible then as it gets *hotter* it revs up, as it stands now, if you don't have a laptop cooler (or mod like mark_ar) there will be NO air flowing through that case! STUPID design Dish!!!
A heatsink would help the chip but you'd totally need a fan to remove the heat off that sink! I think one of those bios heatsink/fans on the ASUS and other neon mobo makers would work great in this situation.
Also bhelms I'm jealous... you you've *obviously* had your 622 for over 7 years now!
A heatsink would help the chip but you'd totally need a fan to remove the heat off that sink! I think one of those bios heatsink/fans on the ASUS and other neon mobo makers would work great in this situation.
Also bhelms I'm jealous... you you've *obviously* had your 622 for over 7 years now!
Actually there are at least 3 Broadcom chips in the 622 and they all run hot. (I'll count them tonight.) The biggest one, which I believe is the decoder, seems to be the hottest. That's the one I tested.abricko said:
Agreed! Maybe they can "fix" it with a SW update, make it more "robust"...?
Yeah, that's my recommendation...
Do you have a link handy?
I was wondering how soon someone would pick-up on that! Obviously the 622 is not the only electronic device that "glitches" once in awhile. But I don't think my camera needs a fan, anyway!...Also bhelms I'm jealous... you you've *obviously* had your 622 for over 7 years now!
(It's a 6 y.o. Kodak 2.2 MP unit, about the size of a brick! And it cost me more than the 622. Since it's in its old age I like to give it a break...)
BRgds...
Major contributor to total overheating is the DISK. As I mentioned in other thread it is dissipate 20+ W into air. Regardless BCM chips temperature, they are do not exceed 10W level total. You'll easy prove it, if disconnect power connector from the disk.
I don't dispute your statement, but the Broadcom chips are dissipating enough power individually to raise their own temps. substantially. The best apparent solution at this time is to get more air moving and that will help ALL heat contributors...Smith said:
I did ask before - is the "E" revision use different disk then I see ? I know if there sit 7Y250S0, temperature lowered for 20 F.
Jim5506 said:
Fan blowing into right side instead of pulling air out of left side: (06/17/2006)
H - 123
L - 89
A - 107
Can somebody explain this? I have a temp. sensor laying on top of the case over the power supply, it transmits to a reciever where I can see the temp at all times. The temp reads 82 degrees right now, has been as high as 89. As you all know you can actually feel a hot area over the power supply. Well a couple nights ago I was getting ready to shutdown the 622 and I felt the case and there was no heat as normal, it was cool, temp was 71, I had the recorder going. Next morning, after being in stand-by all night, it was back to 82 degrees. Any thoughts on this??
See This Post: #149mraroid said:
It explains that I took the power off of the internal power supply conector strip, pins 1 and 2. I had a choice of 5, 7.5 or 12V at the connector.
I figured that this would give the least amount of stress on the rest of the components past this point and not have to worry about the 500ma limit of the USB. (My fan only pulls 140ma)
After the L3.59 software update last night, my numbers moved again...
High: 93
Low: 66
Aver: 78
I have done nothing to change the cooling in weeks now... I can live with these numbers.
High: 93
Low: 66
Aver: 78
I have done nothing to change the cooling in weeks now... I can live with these numbers.
I added a regular PC fan to my rack where it sucks air out of the power supply side of the 622. It's a 12v fan running off of a 9v power cube so that the fan speed and noise are reduced. The temp readings are:
High 107
Low 91
Average 96
High 107
Low 91
Average 96
I added an old PC fan like Mark_AR did connected to pins 1 & 2 of the power supply. But, my wires were not long enough. I instead have it at an angle between the power supply and hard drive where it can blow downward across the board and under the drive. The average temperature indication has dropped 2 degrees so far. This is with the top cover removed. I don't notice any noise from it.
Pepper
Every update, you should see the numbers drop a little more. They seem to be a weighted average of the temps at the time of update.
Remember, the temps are taken from the internal sensor at the base of the motor at the HDD itself.
I was thinking about a [Summer<==>Winter] toggle switch too. I could set the fan to 12v in the summer, and 7.5 or 9v in winter.
Every update, you should see the numbers drop a little more. They seem to be a weighted average of the temps at the time of update.
Remember, the temps are taken from the internal sensor at the base of the motor at the HDD itself.
I was thinking about a [Summer<==>Winter] toggle switch too. I could set the fan to 12v in the summer, and 7.5 or 9v in winter.
which "update" are you referring to, a software update, the nightly guide update, or something else?
the hard drive is cool to the touch, before it could just about burn your fingerprints off.
the hard drive is cool to the touch, before it could just about burn your fingerprints off.
