E-waste sounds like what my girlfriend would call my electronic projects. 
I'm building my amps in the 4U deluxe case at least partially in an attempt to appease her sense of style. I asked if she preferred the plain face or with handles. Her response was that a TV, cable box and Bluray player were enough electronics for our living room. Uh Oh. Guess I have to figure a way to hide the equipment rack and make my speakers look like sculpture.
I'm building my amps in the 4U deluxe case at least partially in an attempt to appease her sense of style. I asked if she preferred the plain face or with handles. Her response was that a TV, cable box and Bluray player were enough electronics for our living room. Uh Oh. Guess I have to figure a way to hide the equipment rack and make my speakers look like sculpture.
Sorry , have a family emergency. Will check this out soon. For now here's the link to the UMS. Could OS and some other people compare OS's board mounting holes to the standard?
The UMS diagram is "upside Down" relative to OS's image.
Link to UMS - http://www.diyaudio.com/forums/imag...sal/universal-mounting-specification-v2.1.pdf
The UMS diagram is "upside Down" relative to OS's image.
Link to UMS - http://www.diyaudio.com/forums/imag...sal/universal-mounting-specification-v2.1.pdf
I used the PDF that went with the badger boards , post a link if there is a "UMS" PDF .
I did move the V+ rail trace (and others) to get more clearance for the bottom holes.
1. Upper (to) lower holes = 60mm (76 - 60 /2 = 8mm from top/bottom)
2 . top holes = 240mm centered (253 -240 /2 = 6.5mm from L/R )
3. bottom holes= 120mm centered (253 - 120 /2 = 66.5mm from L/R)
Holes are 5mm - what do you want ?
Your Email "hates" my Email (I am tagged as a spammer- mailer deamon), if I could just post the sprint stuff here would make it easy. (I have to make a whole new email account - most likely Gmail).
This is good - input !
#1-That would require redrawing the Macro (alexmm's macro) ...easy !!
#2- looked and thought this out .... Q9 forms the high gain pair with Q10 -best kept very close to Q10. With a little lead magic one can mount Q9 much lower than Q10 , the Q10 bolt on mine is 3-4mm above the top of Q9. Mount Q10-12 as high as they will go. (datasheet says 20mm+ without clipping the leads).
#3- You want MORE holes !
I looked at my boards - I will add them (double sided pads - non masked) .
OS
Last edited:
The "5th hole" doesn't line up with the UMS , but maybe it is just a "bonus " for those drilling their own holes. Any extra holes that work with the UMS would have to be either in-line with the bottom holes, or along a line 20mm above the bottom holes
Not required, but just mentioning it, as you asked about extra holes:
If R53 and R54 were moved upward along with F1 and F2 and V- and V+ then maybe there could be a bottom left hole and there is room for a bottom right hole already, inline with the current bottom holes 60mm below the top holes. These would be in addition to the existing bottom holes at the 120mm apart, centered, spacing. The 120mm bottom centered existing holes are the most important because of the shorter distance between them, but nothing wrong with 4 bottom holes. The upper holes in the corners are fine as the active devices help support the board too.
Not required, but just mentioning it, as you asked about extra holes:
If R53 and R54 were moved upward along with F1 and F2 and V- and V+ then maybe there could be a bottom left hole and there is room for a bottom right hole already, inline with the current bottom holes 60mm below the top holes. These would be in addition to the existing bottom holes at the 120mm apart, centered, spacing. The 120mm bottom centered existing holes are the most important because of the shorter distance between them, but nothing wrong with 4 bottom holes. The upper holes in the corners are fine as the active devices help support the board too.
Hope everything is alright , Variac.
The UMS pdf is confusing relative to this amp . For the forum to check the holes , I would have to post the .LAY file (below) . It appears "close" , but further confirmation is better.
I'm down to the resistors on my mouser order - http://www.mouser.com/ds/2/427/rwm-239918.pdf
These are 12mm/ 3W/.22R emitter resistors , they fit the Badger better than the 20+mm ceramic and are a good alternative to the radial "Nobel" resistors originally specified.
Besides referencing EACH part for a perfect fit on both the 2.3 and 2.4 boards
, I standardized the sourcing -
Caps - ALL Panasonic (electolytics) - WIMA (film)
Resistors - all Xicon/Vishay
Semi's - Fairchild /ON semi.
I'll post the pdf and explore how to share my "shopping cart" with others.
Here is the Sprint file -
OS
Save the Mouser cart as a project, then click on share project. They provide a web link near the bottom of the page to allow us to access your order.
Cutco Knives.
Got them 17 years ago for our wedding. Still perfect. I spend a couple hours a month honing the chef's knife and it can split hairs.
Two questions:
As Mouser currently has only one channel available for mjl and mjw transistors and waiting time is over 4 months I consider two options. Buy njw as both P and N are available or take Toshiba 2sa1943 and Fairchild 2sc5200. Reason is that Fairchild's 2sc5200 datasheet shows better linearity than Toshiba's datasheet for 2sc5200. Would that be OK? I need expert opinion.
In the past I mixed irfp240 from IR with Fairchild's equivalent of 9240 as IR's irfp9240 displayed some nonlinearity. It worked judging by ears. I did not investigate how well it worked with test equipment.
The other question is about Q9. This is a pretty important transitor in such circuits. Needs to be fast, have high gain but also good linearity. I think there might be better alternatives than ksa992. Potentially 2sa970bl could be a touch better as it has somewhat better linearity.
In some designs such as Stochino I used 2sa1541e. Not so high Hfe but really good other specs. Again I did not check the difference with lab equipment.
What do you think? What would be the best transitor for Q9?
cheers,
PS I'll probably use 91pF and 430pF for C7 and C8 as I have these unless 100pF and 470pF are truly the most optimal values - ???
As Mouser currently has only one channel available for mjl and mjw transistors and waiting time is over 4 months I consider two options. Buy njw as both P and N are available or take Toshiba 2sa1943 and Fairchild 2sc5200. Reason is that Fairchild's 2sc5200 datasheet shows better linearity than Toshiba's datasheet for 2sc5200. Would that be OK? I need expert opinion.
In the past I mixed irfp240 from IR with Fairchild's equivalent of 9240 as IR's irfp9240 displayed some nonlinearity. It worked judging by ears. I did not investigate how well it worked with test equipment.
The other question is about Q9. This is a pretty important transitor in such circuits. Needs to be fast, have high gain but also good linearity. I think there might be better alternatives than ksa992. Potentially 2sa970bl could be a touch better as it has somewhat better linearity.
In some designs such as Stochino I used 2sa1541e. Not so high Hfe but really good other specs. Again I did not check the difference with lab equipment.
What do you think? What would be the best transitor for Q9?
cheers,
PS I'll probably use 91pF and 430pF for C7 and C8 as I have these unless 100pF and 470pF are truly the most optimal values - ???
Mouser has several pairs of output devices probably usable with Honey Badger:
OnSemi NJW0302, 631 in stock & OnSemi NJW0281, 566 in stock (566 pairs)
OnSemi NJW1302, 788 in stock & OnSemi NJW3281, 241 in stock (241 pairs)
OnSemi MJL4281, 164 in stock & OnSemi MJL4302, 1038 in stock (164 pairs)
OnSemi MJL21194, 15743 in stock & OnSemi MJL21193, 209 in stock (209 pairs)
OnSemi MJ21194, 372 in stock & OnSemi MJ21193, 385 in stock (372 pairs)
OnSemi MJL21196, 50 in stock & OnSemi MJL21195, 72 in stock (50 pairs)
OnSemi MJ21196, 754 in stock & OnSemi MJ21195, 349 in stock (349 pairs)
Toshiba 2SC5200, 5335 in stock & Toshiba 2SA1943, 5091 in stock (5091 pairs)
Fairchild 2SC5200, 394 in stock & Fairchild 2SA1943 1817 in stock (394 pairs)
(total of 7436 pairs)
P.S. Just in case, there is also an odd match, OnSemi NJW1302 with NJW0281.
OnSemi NJW0302, 631 in stock & OnSemi NJW0281, 566 in stock (566 pairs)
OnSemi NJW1302, 788 in stock & OnSemi NJW3281, 241 in stock (241 pairs)
OnSemi MJL4281, 164 in stock & OnSemi MJL4302, 1038 in stock (164 pairs)
OnSemi MJL21194, 15743 in stock & OnSemi MJL21193, 209 in stock (209 pairs)
OnSemi MJ21194, 372 in stock & OnSemi MJ21193, 385 in stock (372 pairs)
OnSemi MJL21196, 50 in stock & OnSemi MJL21195, 72 in stock (50 pairs)
OnSemi MJ21196, 754 in stock & OnSemi MJ21195, 349 in stock (349 pairs)
Toshiba 2SC5200, 5335 in stock & Toshiba 2SA1943, 5091 in stock (5091 pairs)
Fairchild 2SC5200, 394 in stock & Fairchild 2SA1943 1817 in stock (394 pairs)
(total of 7436 pairs)
P.S. Just in case, there is also an odd match, OnSemi NJW1302 with NJW0281.
Last edited:
Project order is done ??
As long as it is high gain -high Vceo - It must endure the full positive rail. BTW- Q9 is the most important semi , with most of the VAS's current gain developed here.
In the Mouser project below (It's done !!) , I chose the KSA992-F grade /200+hfe , best they had. If you can source better , go for it !
http://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=6a73848ab0
With the access ID in the link , members can edit/ or email themselves the project. I have no credit card stuff and my old Tennessee address is listed.
Project is "DIYBADGER1".
It is $54 argggg!! ....
Any errors/ or updates - please share on the thread ....
OS
As long as it is high gain -high Vceo - It must endure the full positive rail. BTW- Q9 is the most important semi , with most of the VAS's current gain developed here.
In the Mouser project below (It's done !!) , I chose the KSA992-F grade /200+hfe , best they had. If you can source better , go for it !
http://www.mouser.com/ProjectManager/ProjectDetail.aspx?AccessID=6a73848ab0
With the access ID in the link , members can edit/ or email themselves the project. I have no credit card stuff and my old Tennessee address is listed.
Project is "DIYBADGER1".
It is $54 argggg!! ....Any errors/ or updates - please share on the thread ....
OS
I simulated with 91/430pf - C7-8 ..... all is well , unity loop gain margin rises by 50k , phase margin OK. Just a slightly "faster" badger with a little more open loop bandwidth. Same general specs as the real life bandwidth is determined by the input RC.
BTW , play around with R24 (TMC feedback) 680R or 1k, TMC is CRC .... changing R24 can compensate for changes in C7/C8.
I have run as low as 47/220pf on my badgers. (wide margin - still stable
)OS
Thanks for the comments on Q9 and TMC and info on transistors. I have about 100 of 2sa970bl and no ksa992.
Yesterday, Mouser Australia had only N and P pairs for NJW, Toshiba and Fairchild. I did not check MJL4xxx. I have 8 or 10 pairs at home but it has worse linearity so it's not my favorite. Unfortunately, only one channel was available for MJL3xxx, MJWxxx.
I considered Toshiba/Fairchild as these are TO-264 so have larger contact area. However, fairchild 2sc5200 has better linearity than Toshiba while Toshiba 2sa1943 has it better than fairchild. At lest that's what datasheets show. So matching two different manufacturers theoretically makes sense here. Mixing these within the same channel could crerate problems and would require larger emitter resistors.
Talking about resistors I'll use 0.1ohm non-inductive as I'm going to get best matches from batches of 25pcs for each channel. With not so well matched laterals i always used 0.1 or 0.15 and with matched bjts as well. As R49 i'll use no more than 1.5-2.2 ohms (5W). Less ringing. R2 will also be small, maximum 220 so C2 will be larger.
Another change I'm going to make is gain. Most my amps have it set at 28 so I can mix them in my active systems. It's also some precaution as I have a teenager at home. Gain over 40 is huge as most preamps can deliver at least 2.5V and most well over 5V - especially if driven from CD/SACD players.
cheers,
Yesterday, Mouser Australia had only N and P pairs for NJW, Toshiba and Fairchild. I did not check MJL4xxx. I have 8 or 10 pairs at home but it has worse linearity so it's not my favorite. Unfortunately, only one channel was available for MJL3xxx, MJWxxx.
I considered Toshiba/Fairchild as these are TO-264 so have larger contact area. However, fairchild 2sc5200 has better linearity than Toshiba while Toshiba 2sa1943 has it better than fairchild. At lest that's what datasheets show. So matching two different manufacturers theoretically makes sense here. Mixing these within the same channel could crerate problems and would require larger emitter resistors.
Talking about resistors I'll use 0.1ohm non-inductive as I'm going to get best matches from batches of 25pcs for each channel. With not so well matched laterals i always used 0.1 or 0.15 and with matched bjts as well. As R49 i'll use no more than 1.5-2.2 ohms (5W). Less ringing. R2 will also be small, maximum 220 so C2 will be larger.
Another change I'm going to make is gain. Most my amps have it set at 28 so I can mix them in my active systems. It's also some precaution as I have a teenager at home. Gain over 40 is huge as most preamps can deliver at least 2.5V and most well over 5V - especially if driven from CD/SACD players.
cheers,