Aleph3 to cold?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi, finally finished my aleph3 clone today! Worked pretty darn good on first power up. (some hum from the transformer that seems to find it’s way to the speakers, or it might be my star grounding being far from perfect. But hey it’s my first amp!) Pictures to come, just have to get them developed. The thing is however that I think it’s running a bit too cold! (imagine that!). I don’t have a suitable thermometer to check the temp on the sinks, but from touching the sinks (0,3c/w /channel) I guess it’s a mere 45 deg. celsius after a few hours of playing music at a moderate level. I can hold my hands there without feeling any pain.
Now, I was a bit concerned about the moderately sized sinks so I did use thermal grease between all the chassis parts, but I don’t think it ads up to that much. I can also touch the case of the transistors for a few seconds or so and it’s slightly discomforting (hot, but not burning hot.) I have to admit I’m a bit concerned about the lifespan of the transistors, don’t want them to go blowing up on me and causing my speakers to blow up or something.
So, you guys think I’m in the clear or what?
A bit about the setup if it helps:
600VA 2x18V transformer, 25,4 volts unloaded, lags down to 24,3 volts under load and seems to stabilize there.
22x4700uf caps
8xIRFP240 isolated with self adhesive silicone pads and thermal grease
Everything else is pretty much right out of the service manual.
 
Posted by Nelson Pass on 01-22-2003 06:18 AM:

Well let's see if your temperature calibration nomenclature
matches mine:

Blimey hot is 10 seconds hands on = 45 deg C.
Crikey hot is 5 seconds = 50 deg C.
Bloody hot is 2 seconds = 55 deg C.
X*?@! is 60 deg C.

All of these are within an acceptable range, although
X*?@! is resrved for the more mature constructor.

This comes from the AlephX thread, but is a nice thumb rule for every amp.
What do I see...you live in Norway! Lucky you, in Africa they have to cool their heatsinks actively, or double their size. :devilr: ;)

Measure the current through the amp.
0.22 10Watt resistors or something, in place of the fuse.
Calculate the power consumption.
Say 150W * 0.3°C/W = 45°C + room temp.

One thing to consider: If memory serves, thermal grease and silpads shouldn't be applied together, but maybe Peter Daniel can tell you more about that. I could be wrong here.

/Hugo ;)
 
Tanks for replaying,
Lets see, circlotron: ambient was no more then 20, so rise is something like 25 deg. above ambient. According to my estimates it should be at least 35 above ambient if I take the chassis dissipation into account.
Netlist: I have searched and I believe you are right about the thermal grease, at least in the way that it’s not necessary. But dos it make things worse?
I’ll try to measure the current drawn this weekend and post back. I read thru Rod Elliot’s article about heatsinks and I think I’ll try to add a aluminium bar or something to help clamp the irf’s better to the heatsink.
Audionut: Thanks for the suggestion , I will certainly consider it. The space is very limited so I’m unsure if I can manage to squeeze it in there. Any idea as to how many volts it will drop?
By the way how much hum would you consider acceptable from a brand new 600va toroid? The hum is audible from a couple meters away, no speakers or input connected, just the hum coming from the transformer.
 
Long ago, I read an article about the advantage or disadvantage of mica vs silpad.
I remember mica and grease is still the best way to transfer the heat, if well applied.
Wonder what Passlabs uses these days.
Nelson must be able to tell us what's best.
Of course grease can be messy.
The same article said that there could be degradation of the silpads over time when grease was used with them.
But it's too long ago and I can't find the article anymore, so I could be wrong here.
Not to scare you of course.

/Hugo ;)
 
audionut said:
wind two 2mh air core inductors out of 150 ft of #14 enamled wire and put them incorporate them with the filter caps, and I bet your hum will go away.


Could i also do this to my aleph 2 that is in progress. ? Would it bennifit me in any way. Im looking for all the improvements that i can do to my amps so that i have the best that i can build : O )
 
Removed the thermal grease today, and mounted a bar across the transistors. Temp on the sinks is now a little bit higher, I believe it’s getting close to 50 deg. C. I believe over-tightening of the screws to be the biggest contributor to the high thermal resistance. The thermal grease on the side of the transistor closest to the screw the grease was almost squeezed away, while on the opposite side there was plenty of grease.

Haven’t gotten around to measure the current as my el-cheapo dmm can’t measure ac current. I did however try to measure dc current on the positive half of the supply, and got some 2,3 amps (both ch).

I also rearranged the star grounding so that its now located at the caps. I think it helped to reduce the hum slightly. Next will be to incorporate a clc filter.
 
:happy1: Making progress. Today I removed R113 altogether, as it turned out the amp was slightly underbiased. I now measure 0.50 volts +- 0.02 across r120-123. From browsing the forum it looks like the use of bc550 as q104/105 results in lower than expected bias.
DC offset it less then 20mV on both channels witch is more than 10mV less then it was before I took out r113. Also put a copper shield around the transformer witch helped to further lower the hum.
About pi filters, would it matter if I wound, lets say 8 x 0.5mH inductors instead of 2 x 2mh? I think it would be easier to fit several smaller inductors than two big ones in to the very limited space I got left. Internal dimensions is only 30*27*10 cm (LxWxH), and with the 22 caps and the 14.5 cm diameter transformer there’s really not much space to spare.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.