adjustment and limits
I am Very impressed by the HB built to spec. No doubt about it.
It is just that there are inside it these little knobs one might adjust...I cannot help but wonder if I can make things a little bit better by adjustment...even if the adjustment is not in the build guide. I Do wonder though what are the limits to safe adjustment.
If turning up the bias is limited only by heat sink and common sense I see no harm in doubling it but I am new and may not see the pitfall...
I am Very impressed by the HB built to spec. No doubt about it.
It is just that there are inside it these little knobs one might adjust...I cannot help but wonder if I can make things a little bit better by adjustment...even if the adjustment is not in the build guide. I Do wonder though what are the limits to safe adjustment.
If turning up the bias is limited only by heat sink and common sense I see no harm in doubling it but I am new and may not see the pitfall...
I built to spec, no issues at all.
I'm just wondering if more bias, keeping it in class A longer, will make it even better ?
if your heatsinks are up to it, you can increase bias to increase idle output stage dissipation for wider class A window, you can do it in small steps and listen each time...
Doubling the output stage bias current adds 6dB to the ClassA output before transitioning to ClassAB........................but wonder if I can make things a little bit better by adjustment...even if the adjustment is not in the build guide. I Do wonder though what are the limits to safe adjustment.
If turning up the bias is limited only by heat sink and common sense I see no harm in doubling it but I am new and may not see the pitfall...
That wil probably make a significant difference to the sound with certain types of speakers and for some specific types of listening levels.
If you find that you prefer the high bias version, then I suggest you look at adjusting the power supply to send much lower voltage to the amplifier.
A similar improvement can be had by doubling the number of output devices for the same doubling of output bias. This keeps the amplifier at the same bias per device.
A combination of doubled devices and reduced voltage would make a good ClassAB with 4 times the ClassA power capability (that +6dB).
I found the gain problem for the one channel. The signal and ground were swapped at the input to the terminal block. As far as bias is concerned I'm running at 50mv from TP1-2. That is around 114ma per pair. After an hour of low level listening the heatsinks were very warm but I could hold my hand on them continuosly without pulling back. My sinks are not that big but also have heavy aluminum angle bracket internal that they are bolted to. It's actually the chassis and power supply of an old Quartro 250 that works out well. Oh yeah this is with MJL21193/4 outputs and +/- 60 volt rails.
Thanks for the input!
AndrewT: your responses are as always, informative and to the point. Thank you!
My only option for dropping rail voltage on this build is to cut it in half to 31VDC. That seemed too low when I put it forth to OS. This does not discount to possibility of a second build
I'm in the process of building a Nelson Pass F5t v2. I'll compare the HB to the F5t and decide which way to jump: HB with maybe 40VDC rails and high bias for class A or a F5tv3. Or maybe a balanced Aleph...I wish I had the chance to visit other audio guys and hear what is going on with their builds. As it is, I am directed by chance to keep on building...not so bad an option
Once again, I thank you as well as the many others lending me their knowledge making my builds possible! It is a good thing we have going here!
Cheers!
AndrewT: your responses are as always, informative and to the point. Thank you!
My only option for dropping rail voltage on this build is to cut it in half to 31VDC. That seemed too low when I put it forth to OS. This does not discount to possibility of a second build
I'm in the process of building a Nelson Pass F5t v2. I'll compare the HB to the F5t and decide which way to jump: HB with maybe 40VDC rails and high bias for class A or a F5tv3. Or maybe a balanced Aleph...I wish I had the chance to visit other audio guys and hear what is going on with their builds. As it is, I am directed by chance to keep on building...not so bad an option
Once again, I thank you as well as the many others lending me their knowledge making my builds possible! It is a good thing we have going here!
Cheers!
F5 and F5t are ClassAB amplifiers.
The difference is that the F5 & F5t use mosFETs and they don't have a optimal ClassAB bias current.
The crossover distortion just gets less as the bias current is increased.
There is no reason to run a F5, nor F5t, at low bias, except to save on heatsink and PSU.
Bias the mosFETs into high bias ClassAB, as high as the heat will allow.
That gets you into lots of ClassA before the amps transition into ClassAB.
A high bias ClassAB that never leaves ClassA should sound virtually the same whether it's a mosFET, or a BJT.
The difference is that the F5 & F5t use mosFETs and they don't have a optimal ClassAB bias current.
The crossover distortion just gets less as the bias current is increased.
There is no reason to run a F5, nor F5t, at low bias, except to save on heatsink and PSU.
Bias the mosFETs into high bias ClassAB, as high as the heat will allow.
That gets you into lots of ClassA before the amps transition into ClassAB.
A high bias ClassAB that never leaves ClassA should sound virtually the same whether it's a mosFET, or a BJT.
101mV
Hello out there.
I thought I'd report that I have taken up the bias on my Honey Badger, TP1 to TP2, to 101mV and yes it does sound better!
Heat sink now gets pretty warm: 110F after an hour but on the Pass forums I get the impression that 130F (55C) is the practical limit. Am I wrong in this?
Cheers!
Hello out there.
I thought I'd report that I have taken up the bias on my Honey Badger, TP1 to TP2, to 101mV and yes it does sound better!
Heat sink now gets pretty warm: 110F after an hour but on the Pass forums I get the impression that 130F (55C) is the practical limit. Am I wrong in this?
Cheers!
in upcoming version of the board, it would be very nice to increase the component hole diameters. I was and am dismayed to see how tiny the holes are, and reading the multiple reports of many components not fitting, seems odd to me that an amp that claims to be "junk part friendly" is hardly that (if only "xyz" component will fit.) That being said, it looks like just the thing to stuff into that broken amp chassis I have in the basement. (Parasound HCA2200). In the mean time, I can see if I can repair the Parasound, while the boards are out. We'll see. Building the HB is a ways away.
Oops!
Hey guys,
So I had both channels working and I started tweaking the bias (R30) on one of the channels. I accidentally cranked it up way too high and subsequently burned R53/R54 and popped the on-board fuses. Oops!
I turned R30 trimmer back to the max resistance, replaced the fuses, and tried to power up again. Now every time I power back up, my mains fuse (6A) blows immediately.
Did I just blow all my output stages?
Visually, I don't see any other components that look out-of-the-ordinary. Nothing else burned. However, if I check continuity between the heatsink and the pins/tab of the output devices, they are shorted at first... after a while, they show like 200 ohms. Not sure what to make of that. I assume this is why my mains fuse keeps blowing now, though.
I also note that TP1 and TP2 are shorted when I check continuity.
What do you think?
Hey guys,
So I had both channels working and I started tweaking the bias (R30) on one of the channels. I accidentally cranked it up way too high and subsequently burned R53/R54 and popped the on-board fuses. Oops!
I turned R30 trimmer back to the max resistance, replaced the fuses, and tried to power up again. Now every time I power back up, my mains fuse (6A) blows immediately.
Did I just blow all my output stages?
Visually, I don't see any other components that look out-of-the-ordinary. Nothing else burned. However, if I check continuity between the heatsink and the pins/tab of the output devices, they are shorted at first... after a while, they show like 200 ohms. Not sure what to make of that. I assume this is why my mains fuse keeps blowing now, though.
I also note that TP1 and TP2 are shorted when I check continuity.
What do you think?
Last edited:
Replace drivers as well, they may not measure badly but are likely to be affected. When I blew some of my output transistors I changed drivers as well. Because I measured all transistors before installing them I could see the difference. Except for the power transistors all the other are cheap.
cheers,
cheers,
Okay, I have everything replaced. I replaced everything on the heatsink for that channel. So outputs, drivers, and the VBE transistor. I meticulously triple checked for shorts between heatsink and all tabs/pins on each of those devices. All looks good.
Cross my fingers and fire it back up. So far so good. Let the amp run for 10-15 min with no issue. Trim off the offset. No issue. Start to trim up R30 again (VERY carefully this time, making sure to go slowly so as not to over-turn it again). So far so good. Get it dialed in right where I want it (to match the other working channel) at 25-26mv across TP1/TP2. I'm within 1-2mv of that, trying to dial in the last bit. Then the 4A on-board fuses die again! No, my hand didn't slip and accidentally short something either.
However, this time, there is no short on my output devices (or any other short to the heatsink). They show dozens of M Ohms. In fact, I can still power it up and the other channel will work fine (no mains fuse failure). It's setup the exact same way... no offset, 26mv across TP1/TP2.
Is it possible that some transient occurred while I'm turning the trimmer, causing some quick current draw?
I'm considering putting the temporary 10ohm resistors back instead of the fuses to compensate, but I wanted to check if that's a good idea before I burn another set of output devices. What do you think?
I haven't touched R7 yet (and I forgot to check it before installing it). Could that be contributing (the CCS adjustment)? It's 200 ohm trimmer, but I have no idea what it's set at right now.
Thanks again!
Cross my fingers and fire it back up. So far so good. Let the amp run for 10-15 min with no issue. Trim off the offset. No issue. Start to trim up R30 again (VERY carefully this time, making sure to go slowly so as not to over-turn it again). So far so good. Get it dialed in right where I want it (to match the other working channel) at 25-26mv across TP1/TP2. I'm within 1-2mv of that, trying to dial in the last bit. Then the 4A on-board fuses die again! No, my hand didn't slip and accidentally short something either
.However, this time, there is no short on my output devices (or any other short to the heatsink). They show dozens of M Ohms. In fact, I can still power it up and the other channel will work fine (no mains fuse failure). It's setup the exact same way... no offset, 26mv across TP1/TP2.
Is it possible that some transient occurred while I'm turning the trimmer, causing some quick current draw?
I'm considering putting the temporary 10ohm resistors back instead of the fuses to compensate, but I wanted to check if that's a good idea before I burn another set of output devices. What do you think?
I haven't touched R7 yet (and I forgot to check it before installing it). Could that be contributing (the CCS adjustment)? It's 200 ohm trimmer, but I have no idea what it's set at right now.
Thanks again!
That is a bit strange as 25-26mV across 0.44 ohm gives only 60mA (you use 0.22 ohm emitter resistors, yes?). Maybe you shortened the output without knowing it.
Nevertheless, start from the beginning, set source current to the ltp as it should be, zero offset and then start to add bias. You should not have problems getting 100mA/44mV between TP1-TP2. If fuses blow again something else is wrong. Anyway, do not exceed say 22mV or 50mA for some time, then go to your current setting of 60mA, then increse bias to about 75 mA. Check temperature of devices/heatsinks, zoebel and zero offset. If nothing happens switcxh off the amp, connect 8 ohm dummy load without any signal and leave it for a couple of hours checking temperature of devices and zoebel resistor and offset.
cheers,
Nevertheless, start from the beginning, set source current to the ltp as it should be, zero offset and then start to add bias. You should not have problems getting 100mA/44mV between TP1-TP2. If fuses blow again something else is wrong. Anyway, do not exceed say 22mV or 50mA for some time, then go to your current setting of 60mA, then increse bias to about 75 mA. Check temperature of devices/heatsinks, zoebel and zero offset. If nothing happens switcxh off the amp, connect 8 ohm dummy load without any signal and leave it for a couple of hours checking temperature of devices and zoebel resistor and offset.
cheers,