Easy? Most likely not.
Possible? Certainly.
Will I try to reinvent the wheel? No.
You are welcome to read every post in the multiple threads. I have read somewhere about the 95 percentile of them over several years time.
Has heatsinking and larger than normal heat sinks been brought up in those past posts? Yes.
Where? Good luck finding. This is an OLD subject / topic with LOTS of information to sort through.
Possible? Certainly.
Will I try to reinvent the wheel? No.
You are welcome to read every post in the multiple threads. I have read somewhere about the 95 percentile of them over several years time.
Has heatsinking and larger than normal heat sinks been brought up in those past posts? Yes.
Where? Good luck finding. This is an OLD subject / topic with LOTS of information to sort through.
Hi Guys,
First let me say thanks to Uriah for taking the time to make this group buy a reality.
I've begun stuffing my boards and have a few questions.
- Are the 1K Dale resistors supplied for R1 and R4 really rated for 1W, they seem small ?
- The BOM call out for Q2/Q3 to be BC546A, but BC546B supplied. Probabily ok but wanted to make sure.
- R11 spec'd as 1R, reading ~ 1.3 - 1.4 R with my meter. I'm assuming it is simply meter error with my cheapy meter.
- The BOM calls for two 75 K resistors For R15 & R16, but I got three in my kit. Is the third 75K just a unintended bonus part ?
Thanks,
PJN
First let me say thanks to Uriah for taking the time to make this group buy a reality.
I've begun stuffing my boards and have a few questions.
- Are the 1K Dale resistors supplied for R1 and R4 really rated for 1W, they seem small ?
- The BOM call out for Q2/Q3 to be BC546A, but BC546B supplied. Probabily ok but wanted to make sure.
- R11 spec'd as 1R, reading ~ 1.3 - 1.4 R with my meter. I'm assuming it is simply meter error with my cheapy meter.
- The BOM calls for two 75 K resistors For R15 & R16, but I got three in my kit. Is the third 75K just a unintended bonus part ?
Thanks,
PJN
A sink that size is pretty small. It also depends on what you mean by 'pretty hot'. If you put it in a closed box with no way to vent I wouldnt expect it to last long. I usually use a P4 sink around 3x3x2. About double what you are using. More than enough for the load they have to drive in my house. 96db speakers make it easy on any amp. If you are running 87db and play it loud you will need larger and larger sinks. With my speakers the P4 sinks rarely got warm. I mean you could hardly tell the difference between them and air temperature.
URiah
URiah
I would definitely like to hear (pun intended) your opinions.
I think this circuit is fully capable of resolving the differences in components.
I have purchased six channels (2 from twisted pair, 4 from Uriah's 1st GB), built four,
, so I couldn't justify another two....
Last edited:
Got my package! Thank you one more time, Uriah. Having a fast look at the contents arise me some questions, which I will ask later, when I get time to look at it closely.
BTW, you made a great effort in packaging it, tons of nearly individual bags and very compact all of it. Pictures promised.
Regards,
Regi
BTW, you made a great effort in packaging it, tons of nearly individual bags and very compact all of it. Pictures promised.
Regards,
Regi
I think those 1k's were actually 2W so you are safe. They LOOOK small but have you seen Panasonic 1/4W
Just a few words about the power ratings resistor manufacturers publish:
There is no free lunch - the temperature rise at a given power will be greater the smaller the resistor. The small resistors are constructed of materials that can withstand very high temperatures, up to 230C in the case of the CPF2 resistor. For printed circuit board mounting you can't run anywhere near the rated power and maintain sane temperatures.
The highest DC voltage for the amplifier should be under +/-35 volts so the resistor will dissipate at most 1/2 Watt. That should produce a temperature rise of 40C according to the datasheet. At a normal room temperature and with allowance for a slightly higher internal enclosure ambient temperature, this produces a reasonable operating temperature. But there is no way you could realistically run 2 Watts dissipation with these resistors. In fact they aren't good for much more than 1/2 Watt the way we are using them.
It came with the kit two types of isolators for the heatsink: One grey band, and two transparent ones. For what purpose? using the one I prefer?
Is the LM318H with the metal can encapsulation isolated? I mean, is any pin tied to the encapsulation? I have measure that NOT, but I would like to be sure.
In the LDR and boards bag, apart from the components, it comes with a dual 100k linear pot. Which, BTW, gives a very pleasant sensation when rotating it, very smooth. It is an ALPS too. What I don't know what it is for is a single gang 5k pot from an unknown vendor. I suppose that is for balancing or something like that. Is there any mounting manual or instructions available? I am afraid of skrewing it up I would be very pleased.
In the next episode, how to decypher the attached paper with numbers and measurements from the LDR's
All in all, a very very very nice packed kit, I wasn't expecting a so PRO grade kit.ñ
Regards,
Regi
Is the LM318H with the metal can encapsulation isolated? I mean, is any pin tied to the encapsulation? I have measure that NOT, but I would like to be sure.
In the LDR and boards bag, apart from the components, it comes with a dual 100k linear pot. Which, BTW, gives a very pleasant sensation when rotating it, very smooth. It is an ALPS too. What I don't know what it is for is a single gang 5k pot from an unknown vendor. I suppose that is for balancing or something like that. Is there any mounting manual or instructions available? I am afraid of skrewing it up
I would be very pleased.In the next episode, how to decypher the attached paper with numbers and measurements from the LDR's
All in all, a very very very nice packed kit, I wasn't expecting a so PRO grade kit.ñ
Regards,
Regi
Thanks Regi,
Grey band and transparent: Both for same - you pick. I had the transparent mica ones for this buy and then Steve at APEXJR threw in the grey band which will do the trick as wel.
LM318H you might try the datasheet. I dont know but I imagine you worry about it touching the Caddock heatsink. I might put some grey band on the top of it to be sure.
The 100k remember is excess that I had on hand and is linear. This was thrown in for free as I had hundreds and no use for it. It will get you running but wont give very smooth volume control. It will change volume rather abruptly. You will want to change to a dual log when you get a chance. 5k is for balance. Goes to 'Series Balance' on the little board. Regi, the schematic you drew represents my board perfectly so you already know how to use it.
Dont be afraid as long as you put the components in the right spot as per your schematic.
A few things to consider when using the LDR board.
1: dial that 200R pot to midpoint and THEN fire up the board. This way you have the protection of half of the pot plus a 68R resistor for each shunt LDR. You can not damage the series LDRs.
2: the paper with the numbers: its basically a measurement sheet of your LDRs. If you have a low 'max db difference' on the lowest measurement then use those for shunt LDRs. If you have a high 'max db difference' on the lowest measurement then use those for series. What is High or Low? Its only relative to the 'max db difference' at the other end of the measurements sheet. Pick the best numbers for shunt. All the numbers are "GOOD" numbers. I wont send out anything that isnt a very good match.
Grey band and transparent: Both for same - you pick. I had the transparent mica ones for this buy and then Steve at APEXJR threw in the grey band which will do the trick as wel.
LM318H you might try the datasheet. I dont know but I imagine you worry about it touching the Caddock heatsink. I might put some grey band on the top of it to be sure.
The 100k remember is excess that I had on hand and is linear. This was thrown in for free as I had hundreds and no use for it. It will get you running but wont give very smooth volume control. It will change volume rather abruptly. You will want to change to a dual log when you get a chance. 5k is for balance. Goes to 'Series Balance' on the little board. Regi, the schematic you drew represents my board perfectly so you already know how to use it.
Dont be afraid as long as you put the components in the right spot as per your schematic.
A few things to consider when using the LDR board.
1: dial that 200R pot to midpoint and THEN fire up the board. This way you have the protection of half of the pot plus a 68R resistor for each shunt LDR. You can not damage the series LDRs.
2: the paper with the numbers: its basically a measurement sheet of your LDRs. If you have a low 'max db difference' on the lowest measurement then use those for shunt LDRs. If you have a high 'max db difference' on the lowest measurement then use those for series. What is High or Low? Its only relative to the 'max db difference' at the other end of the measurements sheet. Pick the best numbers for shunt. All the numbers are "GOOD" numbers. I wont send out anything that isnt a very good match.
The metal case of the LM318H is connected to pin 4, the V- supply. Be sure it does not touch anything else in the circuit.
Ok lets get back to this lm318 I seem to be having. The last picture shows the two lm318. Board on the left is with the ic that works the one on the right is the new one which matches the one that came with the package. The board on the right will just click on and off. If I swap the two ic's around the one on the left will now start to click on and off. Each board will power up one at a time if I only use the ic from the left hand board in picture 3.
Which is telling me that the new lm318 I got is also bad? So I bought 5 and all the new ones are bad? I doubt it, but have been wrong before Now it is hard to read the package on the left hand board but they are distinctively different in there markings. Are thes two different manufactures in here. Please Uriah send me one that looks like the left hand one
Also every one seems to think the sinks maybe too small if thats the case I'll get different ones But for now I may just put some fans on them.
Which is telling me that the new lm318 I got is also bad? So I bought 5 and all the new ones are bad? I doubt it, but have been wrong before
Now it is hard to read the package on the left hand board but they are distinctively different in there markings. Are thes two different manufactures in here. Please Uriah send me one that looks like the left hand one Also every one seems to think the sinks maybe too small if thats the case I'll get different ones But for now I may just put some fans on them.
Attachments
sabre66-
Please take as advice only and not criticism.
1. put the Obigatos on the other side of the PCB's. (Heat sink side)
2. rotate the two PCB's 90 degrees counter-clockwise and mount the LM3886's (insulated!) on the side wall with an aluminum heat spreader between the wall and the chips. That will be sufficient for cooling.
As for the 318 issue, are you able to hear music from each board when the "good" 318 is installed in them? Not just relay clicks but undistorted music? (provided that is what is playing and not an audio book or something)
Please take as advice only and not criticism.
1. put the Obigatos on the other side of the PCB's. (Heat sink side)
2. rotate the two PCB's 90 degrees counter-clockwise and mount the LM3886's (insulated!) on the side wall with an aluminum heat spreader between the wall and the chips. That will be sufficient for cooling.
As for the 318 issue, are you able to hear music from each board when the "good" 318 is installed in them? Not just relay clicks but undistorted music? (provided that is what is playing and not an audio book or something)
Thanks for your reply will get done right away, but I dont understand the heat spreader thing. Some photos I've seen are what look like aluminum U bar if thats all It is thats simple enough. Or I'll mount the chips to copper like in Peter Daniels Patek
Yes I am able to hear music when the "good" 318 is installed in each of the boards.
R39 is on the bottom I had to pull it off it was easier to solder back on the under-side. The issue follows the lm318.
As for heatsinks, I asked this same question last year and Andrew gave me the same advice as he's given here.
I built a CarlosFM LM3886 amp running 35V rails and feeding an 8 ohm load, but being the over engineering type that I am I got heatsinks with a rating of 0.9C/W for each chip.
Do they get warm? Put it this way, I've spent ages with my hand on the sinks trying to work out if it's the chip that's making them warm or just the fact that the ambient temperature is hot and they're absorbing that heat.
Some might say that's overkill, but I call it insurance as I had no prior experience with how hot chip amps run.
Size wise they're about 8 inches wide and 4 inches tall, and the fins come out about 1.5 inches, and not that expensive from Farnell (uk). They're made by Fischer.
For this build I'll be going for 1.2C/w sinks, purely on the grounds of size, but given my experience with how warm (or the lack of) that the 0.9's have got for my kind of listening, then I'm happy that 1.2C/w's will work.
I built a CarlosFM LM3886 amp running 35V rails and feeding an 8 ohm load, but being the over engineering type that I am I got heatsinks with a rating of 0.9C/W for each chip.
Do they get warm? Put it this way, I've spent ages with my hand on the sinks trying to work out if it's the chip that's making them warm or just the fact that the ambient temperature is hot and they're absorbing that heat.
Some might say that's overkill, but I call it insurance as I had no prior experience with how hot chip amps run.
Size wise they're about 8 inches wide and 4 inches tall, and the fins come out about 1.5 inches, and not that expensive from Farnell (uk). They're made by Fischer.
For this build I'll be going for 1.2C/w sinks, purely on the grounds of size, but given my experience with how warm (or the lack of) that the 0.9's have got for my kind of listening, then I'm happy that 1.2C/w's will work.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.