Terry has built a "bigger balls amp" than this one, Leach superamp as well as a Leach Low TIM about this size. The options are the issue here. Poor Terry came here asking which one of two should I pick and I threw in one more. I tried to explain the reasoning behind the options, and encouraged experimentation. 😱
As Terry's shop shows, it's a lot easier to do comparison listening with complete amps rather than swapping parts in a completed amp. I feel for him wanting to make the best sounding Honey Badger, given all the different choices have produced good sounding amps. Pass uses the "crude" cascode reference in his F5T. Leach upgraded that with a zener referenced to ground. OS appears to have listened to both "crude" and the theoretically best versions, and to his ears with this circuit theory won.
As Terry's shop shows, it's a lot easier to do comparison listening with complete amps rather than swapping parts in a completed amp. I feel for him wanting to make the best sounding Honey Badger, given all the different choices have produced good sounding amps. Pass uses the "crude" cascode reference in his F5T. Leach upgraded that with a zener referenced to ground. OS appears to have listened to both "crude" and the theoretically best versions, and to his ears with this circuit theory won.
Thanks guys,
The rest of the parts from Mouser showed up today so I will be filling in boards tomorrow. I'll go ahead and throw a 15V zener in there and be done with it. Today I built two of the Mr Evil/PMI capacitance multipliers. I might try one on this amp with a 25-0-25VAC transformer behind it just to see how it sounds. I also finished my X-BosoZ today. Very nice sounding pre. So now the Badger will get my full attention.
Blessings, Terry
The rest of the parts from Mouser showed up today so I will be filling in boards tomorrow. I'll go ahead and throw a 15V zener in there and be done with it. Today I built two of the Mr Evil/PMI capacitance multipliers. I might try one on this amp with a 25-0-25VAC transformer behind it just to see how it sounds. I also finished my X-BosoZ today. Very nice sounding pre. So now the Badger will get my full attention.
Blessings, Terry
Hi Guys,
I'm finishing up my Badger today and would like an opinion about the outputs. I have a few different ones and wonder which would be best for this amp. I'll be using MJE15033/15032 for the drivers.
Here is what I have.
NJW0281/NJW0302
MJL4281/MJL4302
MJL3281/MJL1302
MJL2114/MJL21193
Thanks, Terry
I'm finishing up my Badger today and would like an opinion about the outputs. I have a few different ones and wonder which would be best for this amp. I'll be using MJE15033/15032 for the drivers.
Here is what I have.
NJW0281/NJW0302
MJL4281/MJL4302
MJL3281/MJL1302
MJL2114/MJL21193
Thanks, Terry
I would choose any of them.
The faster x281/x302 may offer a slightly different performance from the slower 21193/4
The SOA are a bit different and your choice of transformer voltage would have a bearing on what you finally choose.
Have you looked at temperature de-rated SOAR and how load impedance affects the results from modeling with different supply voltages?
The faster x281/x302 may offer a slightly different performance from the slower 21193/4
The SOA are a bit different and your choice of transformer voltage would have a bearing on what you finally choose.
Have you looked at temperature de-rated SOAR and how load impedance affects the results from modeling with different supply voltages?
I have used the 21193/4 pair experimentally. They have less Hfe than the X281/x302 but are far superior in the SOA arena.
I never recommended them as ideal for the badger as one would have to run both the VAS and drivers at a higher current.
In my experiment I used a lower value R36 (33R-47R) , increasing the driver standing current , as well as lowering R27(47R = 12ma VAS).
This may well be my "bassbadger" setup , running 40V rails with 2-3R subwoofer loads .
It can be done with these boards... but the above changes will have to be carefully evaluated and tested. Sonically , for anything besides BASS !! , the x281/x302 or 2sc5200/2sa1943 are recommended.
OS
Last edited:
Hi os - will using 15034/5 as drivers require any circuit changes ? I'm using 281/302s for the output.
Sent from my iPad using Tapatalk HD
Sent from my iPad using Tapatalk HD
Wonderful choice - same Hfe/SOA (600mA+ @60v) as 15032/3 - higher voltage - no issue/no changes !
PS - the previously mentioned changes were relevant only for a lower Hfe OUTPUT pair (21193/4) ,they(the changes) would compensate for EF2 "droop" at high current load 🙂
OS
Thanks Andrew. I went back and studied the schematic and what others have done. I obviously need to break this down so it's easier for me to understand. This is a signal wiring diagram that I believe is 100% correct. From what you are saying, I don't need to run a chassis ground from the pads where the negative speaker post connects, but that it wouldn't hurt to do so?
A couple of quick questions about Q5/6 and Q9. The BOM specifies KSA992. I used the project BOM at Mouser and it gave me 1-KSA992 and 2ea-KSA1015... Looking at the data sheets it looks like VCBO and VCEO are 50v for the KSA1015 and 120v for the KSA992. Collector current is 0.1v higher on the KSA1015. At the voltage/current used in Q5/6/9 I surmise that it won't be a problem. Should I change the KSA1015s to the KSA992? If I keep them the same I am going to use the KSA1015s on Q5/6 and KSA992 on the Q9--any suggestions or am I good with this?
That same Mouser BOM specified quite a few of the resistors at 1/8W vice the 1/4W specified in the DIYAudio BOM...
Any suggestions or help would be appreciated. Thanks!
Jason
That same Mouser BOM specified quite a few of the resistors at 1/8W vice the 1/4W specified in the DIYAudio BOM...
Any suggestions or help would be appreciated. Thanks!
Jason
Last edited:
The cascode sees most of the voltage of the amp's "front end" , they..(Q3-4) must be high Vceo.
Q9 is subjected to almost the whole positive rail , it also must be high Vceo.
Q1-2 , Q5-6 can be almost anything. Ksa1015's are fine , in fact any PNP -25 to->100V high Hfe device. Same for the NPN's (Q1-2) they can be any 25-100V+ high gain device.
Q1-2 and Q5-6 should be roughly matched (DMM at least) , as well as R15-16 , R20-21 (emitter resistors). This will result in almost no DC offset even with R17 (DC offset) at 12 O' clock.
I calculated actual currents for ALL the resistors , most of the 1/8 watter's only see .001-.005W !!
New updated BOM below ...
OS
Last edited:
thanks pete. can't wait to get my boards built. what is the minimum supply rails I can run the hb from ? I intend to test them with a bench supply before fitting to my chassis. it only goes to +-24V
also, i don't suppose anyone in the uk can make/sell me a couple of inductors? ! a reel of wire is £15 :-(
thought i'd ask ! 🙂
also, i don't suppose anyone in the uk can make/sell me a couple of inductors? ! a reel of wire is £15 :-(
thought i'd ask ! 🙂
raymondo, do you have any crossover inductors of suitable gauge laying around you can unwind and rewind? My Badger output inductors will be red laquered.
You may be able to buy a low value inductor of suitable gauge for less than that.
You may be able to buy a low value inductor of suitable gauge for less than that.
Hi bob
Thanks for the reply . Unfortunately I have no inductors lying around at all . I've got stacks of op amps , chips , caps , resistors etc but no inductors or suitable wire .
raymondo,
I have a reel of 1.1mm (c. 0.043") transformer wire I use for inductors.You're welcome to as many metres as you need.
PM me.
Brian
I have a reel of 1.1mm (c. 0.043") transformer wire I use for inductors.You're welcome to as many metres as you need.
PM me.
Brian
One question. I have one hundred or so 100nF/275V polypropylene caps used for mains suppression. They are not small but would fit the board I guess. Could these be used as bypass caps? Zoebel?
I have never used such caps in applications of that kind. For zoebel I prefer Wima poly 400v but I ran out of stock. Zoebel cap should really be non-inductive as well as the associated resistor (still have enough caddocks). I also have some poly 100nF 250V and 630V caps used in tube amps which are good enough for zoebel but I'm just curious about other applications for mains suppression caps.
cheers,
I have never used such caps in applications of that kind. For zoebel I prefer Wima poly 400v but I ran out of stock. Zoebel cap should really be non-inductive as well as the associated resistor (still have enough caddocks). I also have some poly 100nF 250V and 630V caps used in tube amps which are good enough for zoebel but I'm just curious about other applications for mains suppression caps.
cheers,
Those main suppression caps work good , I use them from E-waste all the time. 🙂
V2.2/2.3 only has 7.5/10mm pitch for the zoble. My new V2.4 has 7.5 /10/15mm pitch. You could "jury rig" a 15mm suppression (most are this pitch) , it would work fine.
BTW , a 100V cap would work fine for the zoble - this amp does not oscillate , it might ring a little into a pure capacitive load (piezo tweeter array ... maybe ?).
I saw the worst case ... (forgot the miller cap) , R50 turned brown (smoke) ,zobel cap and the rest of the amp survived. 😱
OS
Thanks ostripper. One more question but not a mainstream one. Got my boards on Friday so I started to study more carefully the design.
Why such high resistor values in led circuits? I must be missing something. The formula I'm using for such resistors is: R = (VS-Vled)/Iled. Diodes from my supplier have Vled about 2.1V for red and 2.6V or 3.5V for blue. Iled is usually 20mA (but there are also small 2mA diodes). 20mA gives me resistor value of about 3.3k for red with VS=63V (rail to the ground). It does not have to burn brightly but you have 22k for red and 100k for blue. Could you please explain?
cheers,
Why such high resistor values in led circuits? I must be missing something. The formula I'm using for such resistors is: R = (VS-Vled)/Iled. Diodes from my supplier have Vled about 2.1V for red and 2.6V or 3.5V for blue. Iled is usually 20mA (but there are also small 2mA diodes). 20mA gives me resistor value of about 3.3k for red with VS=63V (rail to the ground). It does not have to burn brightly but you have 22k for red and 100k for blue. Could you please explain?
cheers,
How bright do you want your LEDs? Reducing the brightness requires significantly reducing the current. Blues either are or seem brighter, so I rarely use more than a mA or 2. At 100K and ~60V input a blue LED will let you know it's on, but not light up the room.