Agreed, I think your right, I am of the old school still using mica pads/grease and then splitting them in two.....Ha.
I'll be doing some shopping on line during the lock down.
Q. any one/type specific
TIA
I finished a FW F6 Yesterday and its been running all day on the bench 50c and just put in the workshop system tonight for a listen, sounding sweet as you like.
Last edited:
Pass DIY Addict
Joined 2000
Paid Member
Check these out: Aluminum Oxide Ceramic Insulators
$17 will get you 25 of them and with a thin layer of goop on each side, they measure just as good at thermal transfer as Keratherm at $2/each.
$17 will get you 25 of them and with a thin layer of goop on each side, they measure just as good at thermal transfer as Keratherm at $2/each.
For those who wish to learn more, Eric conducted a fantastic comparison of thermal interfaces and posted the results here: Experiment: Which mosfet thermal interface material is best?.
Left (good side):
offset 5.00v
R8 8.71v
out bias 4.2v
D1 9.3v
Right (bad side):
offset 35.1v (!!!???)
R8 8.47v
out bias 3.91v
D1 9.2v
So the output bias is really not right. Dennis suggested some other modifications I should make so I will check these first, but regardless of the need for those these measurements substantiate my earlier observations. Thanks again.
Pass DIY Addict
Joined 2000
Paid Member
Thanks for the link, wowbagger! Afterwards, I was thinking I should have dug that out to include with my post. 😉
Just had a read through ...never to old to learn (didnt pick up on it when first posted) Great piece of work Eric.
Dennis,
Looking at the BOM, which I ordered the parts off of, R8 clearly says 500R variable, so that's where the parts error comes from. Not that I intend to use that as an excuse for not checking the schematic, but that's the source of the error. I can put a 1000 ohm fixed in this location to resolve the issue if you recommend that - I have them on hand. That said, the part data sheet says that the range for this part is 10 ohm to 2k, and 1000 is in that range. https://www.mouser.com/datasheet/2/54/3299-776927.pdf
Also, I don't understand if this is the issue why one board is working and the other isn't. In any case, many thanks.
/d
If you follow the BOM, R6 is a 562 Ohm resistor, not a jumper. This gives about 1k with room for adjustment.
It looks like the trimpot in the photo has a 501 resistance code. Per the above datasheet, that's a 500 Ohm trimpot, so the easiest solution is to add R6 from the BOM instead of the jumper.
It may not solve your immediate problem, but at twice the current through the LTP, the dissipation in your jfets will be on the warm side, so you might want to fix it anyway.
Last edited:
something terrible happened when i switched out the resistor. is there a despondent emoji? when powering up the fuses blow after about 2 seconds. i don't even know where to check. any suggestions?
Last edited:
Sorry that you're having some issues. Best advice at the moment may be to step away from it for a bit. A wise ZM once told me to take a tea break and come back with fresh eyes and calm heart. I have done similar things, and am a bit of a noob - so if below is too simple, my apologies. I never want to assume someone's skill level. Just trying to help.
After you come back from your break - and assuming you removed both amp boards from the heatsinks to make your changes...
1) Ensure that your amp boards have EITHER
I recommend the 1k and jumper, but that's me.
2) Blowing some fuses is a good thing vs. magic smoke. It is likely not a huge deal. So, deep breath. Check your fuses. Replace if necessary. Then, without the amp boards attached, do your dim bulb check again. If it doesn't pass the dim bulb - that's odd... but report back. I always do a dim bulb check if I take the boards off the heatsinks or disconnect boards from the PSU for any reason. I also check for shorts - see step 4.
If the PSU passes -
3) Check your unloaded voltages at v+ and v-. Pause here if they're not good and come back to the forum. If they're good move on.
4) Reset your pots on your amp boards. Install each board, but don't wire it yet. Check for shorts to the heatsink on all pins of the FETs. If there are any shorts - change your Keratherm or whatever you're using. Once there are no shorts... move on.
Wire one board and do a dim buib check. Disconnect that board and do a dim bulb check with the other board. If they're both good - wire both boards and do your final dim bulb check. If it passes ... move on.
5) Remove dim buib - deep breath - power up - Set bias and offset. Report results. Hopefully with a big smile.
After you come back from your break - and assuming you removed both amp boards from the heatsinks to make your changes...
1) Ensure that your amp boards have EITHER
the 1k resistor and the jumper at R6 OR
the pot you installed and the 562R at R6
the pot you installed and the 562R at R6
I recommend the 1k and jumper, but that's me.
2) Blowing some fuses is a good thing vs. magic smoke. It is likely not a huge deal. So, deep breath. Check your fuses. Replace if necessary. Then, without the amp boards attached, do your dim bulb check again. If it doesn't pass the dim bulb - that's odd... but report back. I always do a dim bulb check if I take the boards off the heatsinks or disconnect boards from the PSU for any reason. I also check for shorts - see step 4.
If the PSU passes -
3) Check your unloaded voltages at v+ and v-. Pause here if they're not good and come back to the forum. If they're good move on.
4) Reset your pots on your amp boards. Install each board, but don't wire it yet. Check for shorts to the heatsink on all pins of the FETs. If there are any shorts - change your Keratherm or whatever you're using. Once there are no shorts... move on.
Wire one board and do a dim buib check. Disconnect that board and do a dim bulb check with the other board. If they're both good - wire both boards and do your final dim bulb check. If it passes ... move on.
5) Remove dim buib - deep breath - power up - Set bias and offset. Report results. Hopefully with a big smile.
Maybe next weekend, will take the entire faulty board apart, then re-populate it. I have a spare set of matched JFET's, I feel crazy thinking I have to match the MOSFET's ...I have even forgotten how to do it...
Looking at your pictures from a few pages back, it looks like the center pin of the 100K bias pot is not soldered, I checked both top and bottom pics. Is that still the case or did you catch that on your reflow exercise?
Also, where are you measuring the bias voltage and offset? When I setup mine, I measured across any of R16, 17, 18 or 19 and should be about .45VDC across any of those resistors. The offset should be measured from the speaker output to ground, I'm not sure how you would get 35V offset, unless there is a missing ground somewhere, or your meter ground somehow ended up on one of the rails. 😱
Hope this helps, keep us posted.
Cheers
Paul
Last edited:
Thanks Dennis. It is soldered. There are two pads through which that center pin can be soldered: the straight pin layout and the offset pin layout. The resistor I’m using is offset. I didn’t solder the unused pad. As far as measurements I followed the directions Jim outlined in His build guide. I will check them again. Thanks
Okay good luck. One thing I noticed on your boards, some leads are still quite long, and may pose a problem one day if they get bent down, may short to another location, you may want to trim them a bit closer, just something to keep in mind.
Cheers
Cheers
A little jobs for tomorrow, replacing the mica with these.... luckily I have some appropriate sized stand offs.
