I completed this amp a year ago and its not missed a beat since, very reliable. Sound quality far surpasses my previous NAD product and runs a
lot cooler, I don't listen at high volumes.
I did splash out on a Schiit Ascard headphone amp which doubles as a front end pre, very happy. Altogether probably into the amp and pre combo at around $750 Canadian. Very happy with the performance value,
and the satisfaction of assembling myself.
For anyone contemplating building one of these, give it a shot, you will not regret it. Take your time, and let the forum help if required, I did and am
still grateful for the advice during the build.
Again thanks to the forum members that enabled me to acheive this.
You will be even more impressed if you build another one and add a good two way active crossover with cut off frequency ideally about 100Hz above tweeters resonace freq. Crossover between mid and tweeter should be first order then. You will hear the difference straight away and the impovement will becomoe more and move audible as the sound gets louder and louder.
cheers,
There is a comparison on sna forums. A few years ago there was a gtg at a high end hifi dealership in Australia and the honeybadger was taken for a comparison. It would be comparible to maybe a decent 1k-2k amplifier imo.
Although an old post, I'd still like to ask a few questions about it.
I have a pair of Dynaudio Special Twenty Five's that I'd like to use to their full potential, hopefully the Honey Badger will prove to be capable. I suspect it will be. Given the post above, what would be a suitable transformer? I plan to do a dual mono build and I'd like the amp to be able to output maximum power at 4Ohms, though that will probably rarely (if ever) be needed. I like the whole system to be able to handle large dynamics swings with ease. Given a good amp, the speakers should be able to more or less "dissapear". The speakers are two way 8", 4 Ohms minimum, rated at approximately 88dB.
And if at all possible, it would be nice if this amp would perform equally well with other, larger speakers, too.
For the transformers, will I need something in the range of 2x30V - 2x35V? 500VA each? Less?
When I use bridge rectifiers, will I still need the snubber circuit? Determining the appropriate values might be a challenge for me - math wise, and since I don't have an oscilloscope nor a function generator to sweep the transformer coil. I migh have to purchase these, if I want to test and set up the amp properly, though...
Thanks!
Last edited:
can somebody help me with these calculations for the correct transformer voltage?
Power output for HB is stated as about 250W @ 4Ohms
U = sqrt(P.R)
U = sqrt(250.4)
U = sqrt(1000)
U = 31.62 Volts.
Does this mean that my PSU's DC-voltage needs to be 32Volts, or the un-rectified output of the transformer?
In case of the former:
My transformer voltage would be:
32/sqrt(2) = 22.7V -> either 2x20V or 2x25V transformer.
Is this correct?
And:
if I would like to use this amp with different (8Ohm?) speakers, would this still work, or would I need a (much) higher voltage?
I suspect that upping the PSU-voltage to say, 40-45V might be better. But that would lower max output at 4 Ohms, correct? Is it best to look for a trade-off between the two? Or just optimize for my current speakers (which I intend to keep for quite some time).
Power output for HB is stated as about 250W @ 4Ohms
U = sqrt(P.R)
U = sqrt(250.4)
U = sqrt(1000)
U = 31.62 Volts.
Does this mean that my PSU's DC-voltage needs to be 32Volts, or the un-rectified output of the transformer?
In case of the former:
My transformer voltage would be:
32/sqrt(2) = 22.7V -> either 2x20V or 2x25V transformer.
Is this correct?
And:
if I would like to use this amp with different (8Ohm?) speakers, would this still work, or would I need a (much) higher voltage?
I suspect that upping the PSU-voltage to say, 40-45V might be better. But that would lower max output at 4 Ohms, correct? Is it best to look for a trade-off between the two? Or just optimize for my current speakers (which I intend to keep for quite some time).
Last edited:
P = IV = I²R = V²/R for Constant DC voltages and currents.
If you want to work with varying currents and voltages you must use
P = IrmsVrns = Irms²R = Vrms²/R
When the varying currents and voltages are undistorted sinewaves, you can in this special case use
P = IpkVpk/2 = Ipk²R/2 = Vpk²/R/2
If you know the Power (P) and R and want Vpk then you use
P = Vpk²/R/2 and rearrange it to
Vpk = sqrt(P*R*2)
eg
Vpk of the output signal is sqrt(250*4*2) = 44.72Vpk
That is the output at the speaker terminals.
The supply from the PSU will need to be higher by the amount of voltage lost through the amplifier. Allow 3V to 8V for that loss through the amplifier.
The PSU will sag while the amplifier is trying to output 250W. Allow for another voltage loss due to PSU sag. Typically 3V to 8V.
Add those two losses together and you find the PSU quiescent voltage, i.e. 44.7 + ~8V
You need about +-53Vdc, to get to your target of 250W into 4ohms speaker.
To get to that target you will need a very good current capability of both the PSU and the amplifier. Otherwise you will have to jack up the transformer voltage of the transformer.
If you want to work with varying currents and voltages you must use
P = IrmsVrns = Irms²R = Vrms²/R
When the varying currents and voltages are undistorted sinewaves, you can in this special case use
P = IpkVpk/2 = Ipk²R/2 = Vpk²/R/2
If you know the Power (P) and R and want Vpk then you use
P = Vpk²/R/2 and rearrange it to
Vpk = sqrt(P*R*2)
eg
Vpk of the output signal is sqrt(250*4*2) = 44.72Vpk
That is the output at the speaker terminals.
The supply from the PSU will need to be higher by the amount of voltage lost through the amplifier. Allow 3V to 8V for that loss through the amplifier.
The PSU will sag while the amplifier is trying to output 250W. Allow for another voltage loss due to PSU sag. Typically 3V to 8V.
Add those two losses together and you find the PSU quiescent voltage, i.e. 44.7 + ~8V
You need about +-53Vdc, to get to your target of 250W into 4ohms speaker.
To get to that target you will need a very good current capability of both the PSU and the amplifier. Otherwise you will have to jack up the transformer voltage of the transformer.
Last edited:
Wow! That's a quick and comprehensive reply, Andrew. Thanks!
So, I'd be safe with a 35-0-35V transformer. Would 2x500VA for a dual mono configuration be sufficient?
Also, I've been thinking to stretch it's class A window a little, to ca 30Watts. Should be possible, if I've read this thread correctly. Would the heat sink of the 4u deluxe case be large enough to cope with that (and the dual PSU transformers and pcb's, for that matter, plus the softstart and speaker protection boards)? Come to think of it, I'd be better off getting the 5u case.
So, I'd be safe with a 35-0-35V transformer. Would 2x500VA for a dual mono configuration be sufficient?
Also, I've been thinking to stretch it's class A window a little, to ca 30Watts. Should be possible, if I've read this thread correctly. Would the heat sink of the 4u deluxe case be large enough to cope with that (and the dual PSU transformers and pcb's, for that matter, plus the softstart and speaker protection boards)? Come to think of it, I'd be better off getting the 5u case.
No.
a 35-0-35 will give ~+-50.5Vdc
That with a very good build could get you to a 200W or just maybe 210W into 4ohms amplifier.
You will need at least 37-0-37Vac and more likely >38-0-38Vac transformer.
I got >300W into 4r0 from a 40-0-40Vac 1kVA with +-30mF and a 3pair output stage rated for 6ohms, or greater.
I would have needed 4pr for a 4ohms rating and I would expect one channel driven to manage ~310W into 4r0 with my 240Vac supply.
A lower mains suppl, say 220Vac, will produce considerably less maximum output.
a 35-0-35 will give ~+-50.5Vdc
That with a very good build could get you to a 200W or just maybe 210W into 4ohms amplifier.
You will need at least 37-0-37Vac and more likely >38-0-38Vac transformer.
I got >300W into 4r0 from a 40-0-40Vac 1kVA with +-30mF and a 3pair output stage rated for 6ohms, or greater.
I would have needed 4pr for a 4ohms rating and I would expect one channel driven to manage ~310W into 4r0 with my 240Vac supply.
A lower mains suppl, say 220Vac, will produce considerably less maximum output.
One more question. I might be getting ahead of myself here...
I see many commercial devices, like cd-players, that have an output of 0,775V. Is that RMS, or peak-peak? If p-p, do I need to change R2/R3 at the input to accomodate for that voltage difference, or will that be handled by a pre-amp? Is it at all necessary?
According to this calculations, the HB-amp, with +-50V power rail, will need 1,22Vp for maximum output.
I see many commercial devices, like cd-players, that have an output of 0,775V. Is that RMS, or peak-peak? If p-p, do I need to change R2/R3 at the input to accomodate for that voltage difference, or will that be handled by a pre-amp? Is it at all necessary?
A completely different set of requirements. Do the sums and discover just how hot this will get and how many output devices will be needed.
cf the KSA100 which is a full ClassA to 50W into 4r0 and ClassAB ~200W into 4r0.
It has ~900VA 0-37,0-37Vac transformer and 2kW of output devices.
You are looking for an extra 50W of ClassAB requiring higher rail voltage and only willing to give up 20W of ClassA capability.
It will require massive resources !
Hmmm...
I don't think I've seen a 37-0-37V transformer, so that would have to be 40-0-40V one.
I'm not sure what my AC-wall voltage is, I'll have to check that. Anyway, I'm definitely not able to build this amp under ideal lab conditions. I'm confident I can pull off a good build with good thermal coupling. But if I don't manage every last mW of power possible, that'll just have to do. I'll most likely build the PSU as is available in the DIY store, with 40mF per rail per channel. I just have to figure out what to do with the recommended snubbers, how to determine workable/best values.
One more thing: wil this amp run from a floating power outlet (no ground connection). For the time being I live in a very old house, with ditto wiring. I'm just "passing though", so to speak and every piece of audio gear I have, works just fine. I expect no different from the HB, but I'd better be safe and check...
Whoops! Scrap that, then...
Hypothetically: if I were to go through with that, I understand from your explanation I'll need more output devices per channel. Is that possible with the available PCB?
I think it's quite clear by now that I don't know anywhere near enough to re-invent the wheel, so to speak. I find all this very fascinating and I'm learning as I go along. I'm postponing the build, until I feel confident I have enough background knowledge to really know what I'm doing.
So far, thanks a big lot for all your explanantions and help!
Last edited:
fit a earth leakage detector/breaker.
If there is a fault to earth it will isolate and leave the "silent" gear safe to touch.
Fit close rated fusing to each piece of equipment.
These will break the supply fairly quickly, if there is an excess current fault, or user abuse.
This should leave you with a reasonably safe to use kit of audio gear.
If there is a fault to earth it will isolate and leave the "silent" gear safe to touch.
Fit close rated fusing to each piece of equipment.
These will break the supply fairly quickly, if there is an excess current fault, or user abuse.
This should leave you with a reasonably safe to use kit of audio gear.
That is precisely my main problem: not enough knowledge to do the math. Don't know where to start, too. In a previous life I've studied basic electronics and I have some soldering skills, but doing calculations on this amp is far beyond me. I'd like to learn, though.
Run the badger as a class AB - 60-75ma per device. It's not a class A , it
won't perform (or last) as a class A
You will be quite surprised how good it can sound for not being a class A.
In fact , play it loud .... it can do this MUCH better than a class A .
PS - I have a 3 pair BJT output class AB (with "monster" MT-200 devices) ,
all I can get is maybe 5W class A (too HOT !) , but it sounds much better running
at a 100ma (class AB) bias. It will do 2-4R loads better , that's all.
OS