* Overview/features:
Made in Korea sub-1W all tube amp, using a two 12AX7. 1 and 1/2 of the
tubes (three triode sections) are preamp and cathode follower Marshall
style fixed tone stack. I say "fixed" because you only get a volume
knob. The circuit uses fixed resistors instead of bass, mids and treble
pots. ...and it's set up like a Marshall with all controls at "10."
But this design is fairly easy to mod (see below).
The remaining section of V2 is a single ended triode power tube (so this
one IS true class A). Good sized output transformer for surprising
tone. Guys often modify Valve Juniors by using a bigger output
transformer. Not necessary on this one.
The amp says 1W, the manual says 1/4W, and an online comment says 60 mW
before clipping. I haven't measured, but the 60 mW clean and 250 mW
dirty seem about right.
The matching cab is a sealed single 10".
* Technical comments:
Well, here's where there was an initial "glitch" that would've been a
headache for MOST owners. The amp went dead in the first few minutes of
play.
When everything goes dark (no pilot, no tube glow, no sound), the
average owner would check the mains fuse (it was fine), then have to
seek warranty service. But I went looking for a filament fuse.
Sure enough, the filament fuse (located on the circuit board) was
popped. When the 500 mA value JUST SEEMED WRONG, I got on the phone to
Crate/Blackheart. The tech support guy was nice enough, but NO HELP at
all. Then I took a second look and noticed that the high tension fuse
socket had a 1A in it. It's obvious that somebody at the factory put
the two fuses in the wrong holes. (See comments below about my phone
conversation with Pyotr Belov, designer of the Blackheart amps, and the
Epiphone Valve Junior.)
The amp comes supplied with two JJ ECC83S tubes. Blackheart/Crate
apparently cared enough to ship them to Korea in favor of the cheaper
and closer Chinese option.
Build quality is VERY GOOD on both the head and cab. The board quality
and component quality exceeds that of the Epiphone Valve Junior. I'll
try to post links to pics before too long.
Both the head cab and the sealed 10" speaker cab are NICELY done in 15
ply 18mm void free plywood (including the back speaker panel, which is
often cheaper material). I'm not sure why they bothered on a sealed
cab, but the interior is sprayed black. The 10" OEM speaker is built by
Eminence, with a purple sprayed stamped steel basket, good sized magnet,
and 30W rating. Again, I don't know why they do this in a sealed cab,
unless they use the same speaker in an open combo.
They used a very nice salt and pepper vintage style grill cloth. The
UGLY Blackheart logos make an otherwise vintage looking setup appear to
be some sort of kiddie toy. I was very happy to find that the awful
logos were easily removed by pulling two screws.
* Subjective tone comments:
Not much clean headroom, but who buys a sub-1W amp for clean? You buy
it for tube distortion at reasonable volume levels.
Many Valve Junior owners end up with way more volume than they bargained
for. After all, 5W can supply half the volume of a 50W amp.
The KA (Killer Ant) is NOT "too loud." It's the perfect volume for a
home owner to use at night, and for an apartment/condo dweller to use in
daylight hours. The wife won't be happy if she's trying to watch TV in
the same room, but in a separate room with closed doors, it shouldn't
bother her -- or wake up the kids. You should never have the cops or
angry neighbor at the door. Yet there's still plenty of volume to get
the feel of a tube amp, and more than enough for recording purposes.
Tonal balance is better than the stock Valve Junior, which tends to be
dark in stock dress. But my tastes say that the KA has a bit too much
bottom end, which leads to a slightly mushy distortion in the lower
frequencies. I solved that by tacking a 33k resistor on the 1M stock
bass resistor. (Remember, this thing has a Marshall tone stack with
fixed resistors for an "all tens" setting). I plan to do more
experimenting with different tone settings. I'm thinking that it
deserves a treble pot, at least. Heck, maybe I can shoehorn in three
tone pots!
* Tube swapping:
In stock dress, the amp generates a very passable "gainy" style tube
distortion. More like a cranked Marshall than an overdriven Fender.
I've experimented with two other V2 options (V2 simply refers to the
second tube in the circuit).
A NOS Sylvania 12AT7 generated a noticeably more power and headroom,
with a gainy distortion that has some edge to it.
A NOS Sylvania 12AU7 also produced quite a bit more power, but even more
controllable headroom, and a "smoother" style of distortion. It "opens
up" the amp, giving a subjectively wider frequency response with both
tighter and more solid bottom end and brighter highs. It also gives the
amp more "touch sensitivity," allowing you to get semi-clean at volume,
and distorted with a little more string pluck. Put another way, it
cleans up nicely by turning down the guitar volume slightly. More like
a Fender than a Marshall.
(Note: to compensate for different tube characteristics, I chose to
intentionally mismatch the 12AT7 and 12AU7 by using a 4 ohm cab in the
16 ohm jack. See comments below from the designer, he doesn't bother
with the mismatch.)
* Tech support drops the ball, Pyotr Belov picks it up
I mentioned the bone-head fuse issue, and how tech support never got
back to me. After looking over the very annoying website, I found the
designer's E-mail address. I fired off a message to him, and shortly
after he sent me a message saying he'd call after he got out of a meeting.
Sure enough, a couple hours later, he called me to confirm the fuse
values. We also discussed tube swapping options. He also likes a 12AU7
for more headroom, although he said he doesn't even bother with doing
the intentional impedance mismatch. (I think doing it gets a bit more
headroom). He said it's now his main playing rig, into a 4x12. I'm
liking mine into a Vintage 30 2x12. It doesn't seem to wake up my 70's
Hiwatt 4x12 enough, but I haven't tried my 80's Hiwatt (more efficient
speakers, so it may work better).
I say that the website is ANNOYING, and so are the owner's manual and
logos. This is a GREAT little amp, but they seem to be marketing it for
8th graders! C'mon, Blackheart/Crate, I think you're focussed on the
wrong demographic. Mature guitar players are the ones looking for tube
distortion at low volume.
*** Conclusion ***
This is a great little amp for saturated tube tones at low volume. But
it's even better if you calm down the bottom end response and drop in a
12AU7 for more touch sensitivity. Or a 12AT7 for an edgy distortion
with more output.
If you need an amp that will do true tube distortion tones at low
volume, this is the one. I had to put an L-Pad attenuator on my Valve
Junior to get the volume down. This one has the right volume levels.
Price is reasonable, especially since it is NOT Chinese. Everything
about this amp seems to be a notch or two higher than Blackheart
could've gone.
Now THAT's a review. Thanks.
On Mon, 07 Sep 2009 14:46:26 -0700, Jim <j...@askmebeforeyousend.com>
wrote:
That kit is a simple tone, some sort of gain boost (which I think would
be counter productive), and a line out.
The simple tone control would just be a passive highs cut.
If I do anything, I'll replace the fixed resistors with pots, so that I
get a real tone stack.
I did my own line out design on my Valve Junior mod, but never really
use it.
So the BitMo kit doesn't have any appeal to me.
Maybe I ought to put together at bass/mids/treble tone stack kit!
Confirmed for the 15 watt version as well. It's a real piece of work.
Handles pedals really well even without an effects loop. There are
times when I ask myself what the hell I bought my Statesman for (but
then I play on the statesman for a while.. ;-)
-
On Mon, 07 Sep 2009 23:08:30 -0700, Jim <j...@askmebeforeyousend.com>
wrote:
This is an ambush, so apologies.
This thread got me thinking about the Epi VJ. The mod site I have found
suggests a switchable 68 or 100 pF cap in the vacant C3 slot as a bright
control. If I wanted a variable increase in brightness, what would be a good
choice of cap and pot?
Thanks.
Tony D
Tony,
There's a lot of lead used that way. I did a bright cap on a switch
right off of the volume pot.
If you wanted to vary the brightness, you could use different value
bright caps. The lower the value, the more "sparkle" (higher shelf
frequency). The higher the value, the more it'll dip towards the mids,
resulting in a sense of more treble instead of just more sparkle. But,
as I'm sure you already know, as you increase the volume level, you
decrease the effect of the bright switch. That's the way I *want* it,
because I want cleaner tones to be brighter, and distorted tones to be
darker.
I use an 82p silver mica. You could get a SPDT "on off on" toggle and
do 82p and 150p.
Is there such a thing as a variable capacitor (not a flux capacitor, which
caused me to be here in the first place), or is it usually done by altering
the ratio of capacitance to inductance via a pot (as opposed to being
equipped with a selection of capacitors and a soldering iron)? It's a
question from my safe standpoint of ignorance.
--
Lawrence
"If I don't see you in the future, I'll see you in the past" - Tommy Bolin -
Melbourne 1975
>As for bright switches on VJs lowering R6 to 100k and having a switchable
>bypass cap over that resistor is a popular mod. I used an old
>polystyrene 1000pF.
>
>Like this:
>http://www.flickr.com/photos/22166118@N05/3807631058/sizes/o/
Yowie, How did the 33nf cap get wired like that? It looks like a
mistake in the drawing. Wired as-is, it won't just notch the mids--it
will take off all the highs as well. Flipping the switch to put the
1000pf in-circuit will restore them -slightly-, but that's going to be
a very bass-heavy circuit. Also, that will greatly subdue the effect
of the treble-cut pot.
Have you ever been inside of an older radio chassis? The odd looking
tuner contraption with alternating plates of metal that swing into each
other is actually a variable capacitor.
But there are easy ways, and ways that don't make sense.
The easy way is to either experiment with a capacitor across the volume
pot, or capacitors across resistors in the circuit. When capacitors are
parallel, they add value. When series, the value gets lower (half, if
both are the same value going in). So with a couple of caps and the
right switch or two, you can have several options.
NEVER open a tube amp chassis unless you understand that filter
capacitors can store HIGH VOLTAGE even when the amp is unplugged!
KNOW THE SAFETY ISSUES before you start experimenting.
Thanks for the suggestions. Jim, you put a treble bleed across the hot and
wiper on the volume pot? That sounds like a good plan to me, or I could try
the dpdt on-off-on option with two caps in the C3 slot. - I'm surprised I
didn't think of that myself, as I have that set up as a tone control on a
Baggs M1 Active soundhole pickup.
Tony D
>Hmmm... The drawing could be wrong. Will check it carefully tomorrow
>night. It's just a rearrangement of this.
>
>http://www.sewatt.com/files/sewatt/Mid%20Notch%20pot%20schem.png
Hi Yowie,
I got a brief look at the link above before it stopped working.
The circuit appears like that of your previous schematic, but the
chart looks like it ignores source impedance. That impedance would
include both the output of the first tube stage (the 12ax7's plate
resistance in parallel with its load resistorr), plus the following
resistor (100K in your circuit, but I believe it's 1M in the original
VJ circuit).
The 33nf (.033uf) cap will work against the source impedance to roll
off most of the mids and highs. So something is odd.
>I even did a quick and dirty simulation using this thing:
>http://www.falstad.com/circuit/
>... because I didn't think it would work - but it did.
I''m not sure if your own model is online (is it?), so I don't know
how you're modeling the source impedance. You could use a 130k
resistor in front of the circuit to simulate it. Actually, you
already have a 100k resistor there in the actual circuit, so if you
included it, that should have shown you the rolloff effect.
Can you do an 'export' (under the File menu) and post the component
net?
You are very correct. The "electrolyte" in that case is the air itself.
Well spotted!
> But there are easy ways, and ways that don't make sense.
>
> The easy way is to either experiment with a capacitor across the volume
> pot, or capacitors across resistors in the circuit. When capacitors are
> parallel, they add value. When series, the value gets lower (half, if
> both are the same value going in). So with a couple of caps and the right
> switch or two, you can have several options.
>
Almost exactly like resistors, except the complete opposite.
> NEVER open a tube amp chassis unless you understand that filter capacitors
> can store HIGH VOLTAGE even when the amp is unplugged!
>
> KNOW THE SAFETY ISSUES before you start experimenting.
>
This is something that has always confused me. People say "don't touch!",
when the better option would be to gracefully discharge the offending
capacitors.
--
Lawrence
"If I was a towel, why would I be wearing this hat and this fake
moustache" - Steven McTowelie - 19 April 2006
Yes, just like a bright switch in typical Fender tube amps.
That's a part of knowing the safety issues.
Agreed, Rich, but discharging the filter caps should be part of a safety
checklist rather than issuing a blanket "Watch out!" warning.
Having said that, what would you think would be a suitable resistance and
power rating to discharge a cap with 400V just waiting for an inquisitive
enthusiast to stumble by (and perhaps not stumble away)?
I have a blender that I make banana smoothies with. After I unplug it, I
always stick a knife across the prongs of the power cord. Sometimes there's
a big spark and sometimes there isn't. I'm presuming that it's dependent
upon where in the 50Hz (or 60Hz in your world) cycle that the appliance
happened to be at the time that I switched it off. Is the same true for
filter caps in an amplifier, or will they always carry a potentially lethal
potential?
--
Lawrence
"You can just hang outside in the sun all day tossing a ball around, or you
can sit at your computer and do something that matters." - Eric Cartman - 4
October 2006
Well, *I'm* not giving shorthand lessons on how to discharges caps in a
newsgroup. ...for fear somebody won't do it right!
Hence my warning.
>
> Having said that, what would you think would be a suitable resistance and
> power rating to discharge a cap with 400V just waiting for an inquisitive
> enthusiast to stumble by (and perhaps not stumble away)?
>
> I have a blender that I make banana smoothies with. After I unplug it, I
> always stick a knife across the prongs of the power cord. Sometimes there's
> a big spark and sometimes there isn't. I'm presuming that it's dependent
> upon where in the 50Hz (or 60Hz in your world) cycle that the appliance
> happened to be at the time that I switched it off. Is the same true for
> filter caps in an amplifier, or will they always carry a potentially lethal
> potential?
I'm sure you're pulling our leg on the above.
To answer the question, many amps do have resistors in the filter
section that will safely discharge the caps in under two minutes. But
you should never count on it. If a resistor goes open (it happens), you
could ruin your day. And some amps don't have them to begin with.
Good point. It would be a shame if someone electrocuted themself while
trying to "make it safe" when they otherwise may not have gone near the
thing!
>>
>> Having said that, what would you think would be a suitable resistance and
>> power rating to discharge a cap with 400V just waiting for an inquisitive
>> enthusiast to stumble by (and perhaps not stumble away)?
>>
>> I have a blender that I make banana smoothies with. After I unplug it, I
>> always stick a knife across the prongs of the power cord. Sometimes
>> there's
>> a big spark and sometimes there isn't. I'm presuming that it's dependent
>> upon where in the 50Hz (or 60Hz in your world) cycle that the appliance
>> happened to be at the time that I switched it off. Is the same true for
>> filter caps in an amplifier, or will they always carry a potentially
>> lethal potential?
>
> I'm sure you're pulling our leg on the above.
>
I'm not actually. I'd been shocked by it on several occasions, so I figured
that I should discharge it intentionally rather than unintentionally. While
I haven't actually investigated, I assume that it's only a relatively small
cap. That would explain why it's only an uncomfortable jolt at best. I'd
never in my wildest dreams expect to still be posting if I tried the "knife
trick" on a filter cap in an amplifier. I may be very silly but... in fact,
it would probably be more accurate to just leave it there!
> To answer the question, many amps do have resistors in the filter section
> that will safely discharge the caps in under two minutes. But you should
> never count on it. If a resistor goes open (it happens), you could ruin
> your day. And some amps don't have them to begin with.
>
Ultimately, you'd suggest that we should all keep clear unless we genuinely
know what we're doing. On reflection, that is very sound advice.
There are some web sites you could check...personally, despite following
the amp group for a couple of years now and understanding the underlying
technical issues, I still don't feel comfortable with it. 400-500 V is
an awful lot of juice and those filter caps can store a considerable
amount of charge. There's also a "memory effect" with electrolytics so
if you measure the voltage immediately after discharge you'll get close
to zero but a spontaneous voltage can build up over time even when
they're disconnected from a circuit.
> I have a blender that I make banana smoothies with. After I unplug
> it, I always stick a knife across the prongs of the power cord.
> Sometimes there's a big spark and sometimes there isn't. I'm
> presuming that it's dependent upon where in the 50Hz (or 60Hz in your
> world) cycle that the appliance happened to be at the time that I
> switched it off. Is the same true for filter caps in an amplifier,
> or will they always carry a potentially lethal potential?
The filter caps see a DC voltage with ripple, and it doesn't matter at
what portion of the cycle they're disconnected from the supply, they
still retain charge. As Jim said, most *modern* amp circuits have
discharge resistors built in but (a) verify for yourself that the amp
you'll be working on has one and (b) don't assume it's working
correctly!
>On Wed, 09 Sep 2009 15:52:38 -0400, RS wrote:
>
>> On 09 Sep 2009 12:24:36 GMT, yowie <yo...@thepub.lan> wrote:
>>
>>>On Wed, 09 Sep 2009 05:56:06 -0400, RS wrote:
>>>
>>>> On 09 Sep 2009 08:07:19 GMT, yowie <yo...@thepub.lan> wrote:
>>>>>...
>>>>>http://www.flickr.com/photos/22166118@N05/3807631058/sizes/o/
>>>>
>>>> Yowie, How did the 33nf cap get wired like that? It looks like a
>>>> mistake in the drawing. Wired as-is, it won't just notch the mids--it
>>>> will take off all the highs as well. Flipping the switch to put the
>>>> 1000pf in-circuit will restore them -slightly-, but that's going to be
>>>> a very bass-heavy circuit. Also, that will greatly subdue the effect
>>>> of the treble-cut pot.
>>>
>>>Hmmm... The drawing could be wrong. Will check it carefully tomorrow
>>>night. It's just a rearrangement of this.
>>>
>>>http://www.sewatt.com/files/sewatt/Mid%20Notch%20pot%20schem.png
>>
>> Hi Yowie,
>>
>> I got a brief look at the link above before it stopped working. The
>> circuit appears like that of your previous schematic, but the chart
>> looks like it ignores source impedance. That impedance would include
>> both the output of the first tube stage (the 12ax7's plate resistance in
>> parallel with its load resistorr), plus the following resistor (100K in
>> your circuit, but I believe it's 1M in the original VJ circuit).
>
>Yep 1M followed by a 1M to ground (R7) in parallel with the volume pot,
>which was also 1M originally.
>
>> The 33nf (.033uf) cap will work against the source impedance to roll off
>> most of the mids and highs. So something is odd.
>
>
>>>I even did a quick and dirty simulation using this thing:
>>>http://www.falstad.com/circuit/
>>>... because I didn't think it would work - but it did.
>>
>> I''m not sure if your own model is online (is it?), so I don't know how
>> you're modeling the source impedance. You could use a 130k resistor in
>> front of the circuit to simulate it. Actually, you already have a 100k
>> resistor there in the actual circuit, so if you included it, that should
>> have shown you the rolloff effect.
>>
>> Can you do an 'export' (under the File menu) and post the component net?
>
>I wget-ed the whole simulator thing to my hard disk and ran it from there
>but I don't appear to have saved the model I made - at least not so I can
>find it again. I did include the 100K resistor and I think 1M to ground
>for the output IIRC. Looking at some drawings I made at the time the
>original 1M R7 was probably still in place initially but at some point I
>decided to remove it. This was 6 months ago and I'm sure not exactly
>what I did.
Hi Yowie,
Any way to duplicate your previous circuit? I'd love to see how that
can -not- eliminate most of the high frequencies.
>I'm getting curious. I never did try to measure the effect on the
>frequency response of the actual amp but I might try that on the weekend.
>
>I should be able to get a fair idea with a software signal generator
>(don't have a real one) feeding the input and a scope across the speaker
>terminals?
I double-checked on one of the circuit emulators that I've written.
The configuration didn't match perfectly, but I swapped high
resistance values for open circuits, etc. The result should be close,
since the overall response is swamped by the effect of source
impedance (100k+) against the .033uf to ground.
The low end looks much like the bass side of a Fender tone stack, but
due to the source-imp/.033uf with no separate treble path (the 250p
cap in the Fender circuit), the response curve continues to dip as
frequency rises. Down -30 db at around 1500 Hz. I believe that is
representative of what you'd see from the circuit above.
I tried to duplicate that on the Duncan Tone Stack Calculator so that
you could see it, but unfortunately that doesn't allow tweaking
individual component values. Sorry that I can't send my emulator
program, but it has always proven an accurate match when I've used it
for circuit designs. (In this case, though, the loading is pretty
obvious from just looking at the circuit)
Anyway, -30db is significant. I have no idea how you're getting treble
through that circuit if it matches that schematic.
Theoretically, a capacitor can never be fully discharged. "Diminishing
returns" springs to mind even if that term is generally not associated with
electronics.
The "memory effect" that you mentioned is a strange phenomenon, and akin to
a battery not working now but having a sufficient charge a bit later on. If
a filter cap was to be discharged manually, it's probably a very good idea
to repeat the process a couple of times, and especially before working in
the general area of the huge blue cylindrical thingy.
>
>"RichL" <rple...@yahoo.com> wrote in message
>news:gJOdnbabPsq0yzXX...@supernews.com...
>> Lawrence Logic <lawrence-Omit...@amd-p.com> wrote:
>>> "Jim" <j...@askmebeforeyousend.com> wrote in message
>>> news:5aqdnc78b6Z-ajrX...@posted.isomediainc...
>>>>
>>>> NEVER open a tube amp chassis unless you understand that filter
>>>> capacitors can store HIGH VOLTAGE even when the amp is unplugged!
>>>>
>>>> KNOW THE SAFETY ISSUES before you start experimenting.
>>>>
>>>
>>> This is something that has always confused me. People say "don't
>>> touch!", when the better option would be to gracefully discharge the
>>> offending capacitors.
>>
>> That's a part of knowing the safety issues.
>>
>
>Agreed, Rich, but discharging the filter caps should be part of a safety
>checklist rather than issuing a blanket "Watch out!" warning.
>
>Having said that, what would you think would be a suitable resistance and
>power rating to discharge a cap with 400V just waiting for an inquisitive
>enthusiast to stumble by (and perhaps not stumble away)?
Hi LL, Try a 220k 2 watt resistor. Caps' discharge curve is
exponential, but it will take enough off quickly to keep you alive.
And if you forget to disconnect it and you power the amp up again, the
resistor itself is high enough in value and wattage that it won't blow
up, burst into flames, or otherwise put a serious drain on your
supply.
If the power tubes are plugged in and warmed up, they'll present a low
impedance that drains the supply pretty fast on power-down. But do
-not- count on that. If you unplug the power tubes, or power up just
long enough to charge the caps, but not heat the power tubes, the
supply will be hot (can't count on preamp tubes to discharge the
supply).
Rich L. is exactly right about the filter caps bouncing back after
discharge. Shouldn't be enough to put you on the floor, but you could
definitely feel it. (Rich is modest, but very good at this stuff,
evidently).
>I have a blender that I make banana smoothies with. After I unplug it, I
>always stick a knife across the prongs of the power cord. Sometimes there's
>a big spark and sometimes there isn't.
You must have an all-tube blender. (That's a bit odd there, Lawrence)
Thanks, RS.
>>I have a blender that I make banana smoothies with. After I unplug it, I
>>always stick a knife across the prongs of the power cord. Sometimes
>>there's
>>a big spark and sometimes there isn't.
>
> You must have an all-tube blender. (That's a bit odd there, Lawrence)
>
It's a Sunbeam, made in China.
A Japanese-made Akai 4000DS reel-to-reel tape deck used to do the same thing
ie 240V sitting across the active and neutral wires of the power cord. It
was only ever a tickle, but it was a BIG tickle.
Maybe I'm more susceptible to shocks than other people. I used to get
zapped when I'd get out of the car, so I now hold the bodywork as I put my
feet on the ground. Problem solved, because the point of discharge is my
shoes rather than my fingers. It's the same when I walk across a carpet,
but I now place a flat hand on the wall prior to touching the door knob. I
discharge at a much more "user-friendly" rate through the paint on the wall.
Of course, those examples are static electricity. You won't hear a static
discharge below around 3000V, and you usually won't feel it below 5000V. A
decent zap from a door knob will generally be around 15000 to 20000V. If
you discharge a couple of hundred volts, you won't even know. That's why
most people are ignorant of taking precaution when handling static-sensitive
devices. Applying 500V instead of 5V, irrespective of the current, doesn't
sit well with discrete components however.
--
Lawrence
"They're French, so they surrendered immediately!" - NATO Captain - 22 April
2009
LL, I should also add that 220k/2W is appropriate if you connect it
with jumpers, hold it there for a bit, or even solder in place
(simplest). If you're looking to touch a probe to B+ for a quick
one-second zap, you'd have to use a lower value resistor to get the
voltage down to safe levels that fast, but then all the caveats above
apply (re disconnects, burnt resistors or supplies, etc). And you
don't want to repeat quick-discharge cycles to an electrolytic.
Re the blender: Many larger motor circuits use a 'start cap' to help
boost torque during startup. It's usually taken out of circuit by a
centrifugal switch as motor speed ramps up. Maybe some mini version in
small appliances. My thing is mainly tube amp circuitry and
microcontrollers. Blenders, not so much. <g> I'll have to trust you
on that.
> I tried to duplicate that on the Duncan Tone Stack Calculator so that
> you could see it, but unfortunately that doesn't allow tweaking
> individual component values.
The version I have does... 1.3 is the version, probably old.
You have to double-click the component and a box pops up.
Any component shown is tweakable.
__
Steve
.
Thanks for the heads up, Steve! The new version has some new models as
well. Not sure if the older one accepted component changes, now that
I think of it. The Duncan program is very cool, but I never had much
occasion to run it (already had my own software written when he put
that up).
Anyway, I tried to simulate the circuit at:
http://www.flickr.com/photos/22166118@N05/3807631058/sizes/o/
by doing the following to Duncan's Fender model:
Zsrc = 130K (sim of (plate res | load res) + 100k series resistor)
Treble cap, C1 = 1uf (virtual wire)
Slope res, R1 = 1 ohm (virtual wire)
R3 (bass) = 20M (virtual open)
C2 = 3.3n
C3 = 33n
R4 (mid) = 3.3k
R5 (load/vol pot) = 500k
Turned up all pots to max.
Comments welcome on that, if you see a better way to model it.
The main difference in topology is that C2 on Duncan's circuit
connects to the opposite side of C3, compared to the sim. But due to
the 1:10 ratio, that should not have a huge effect.
The posted VJ 'mid cut' circuit is much like a standard Fender tone
stack with a shorted slope resistor and shorted treble cap, so it
wasn't difficult to imagine the result. The killer is the source
impedance working against the .033, of course. That swamps everything
else, including the effect of the treble control. The source imp. must
have been zero'd on the models that look like they don't lose treble,
since there's nothing but the little 3.3k resistor to prevent rolloff.
Anyway, the curves from the Duncan program appear identical to those
from my program. The response is down about -30db at 2k. The mystery
is why some report that the circuit works OK. Must be component value
errors or problems with the diagram? And the pot still doesn't work
as intended on the models, even with lowered source impedance.
Yup, the charge (and voltage) decay exponentially. Get it below 1 V and
you'll be fine.
> The "memory effect" that you mentioned is a strange phenomenon, and
> akin to a battery not working now but having a sufficient charge a
> bit later on. If a filter cap was to be discharged manually, it's
> probably a very good idea to repeat the process a couple of times,
> and especially before working in the general area of the huge blue
> cylindrical thingy.
Yup.
>On Thu, 10 Sep 2009 16:52:29 -0400, RS wrote:
[re Duncan tone stack model for the Epi VJ circuit]
>> Anyway, I tried to simulate the circuit at:
>> http://www.flickr.com/photos/22166118@N05/3807631058/sizes/o/ by doing
>> the following to Duncan's Fender model:
>>
>> Zsrc = 130K (sim of (plate res | load res) + 100k series resistor)
>> Treble cap, C1 = 1uf (virtual wire)
>> Slope res, R1 = 1 ohm (virtual wire)
>> R3 (bass) = 20M (virtual open)
>> C2 = 3.3n
>> C3 = 33n
>> R4 (mid) = 3.3k
>> R5 (load/vol pot) = 500k
>> Turned up all pots to max.
>> ...
>> The main difference in topology is that C2 on Duncan's circuit connects
>> to the opposite side of C3, compared to the sim. But due to the 1:10
>> ratio, that should not have a huge effect.
>>
>> The posted VJ 'mid cut' circuit is much like a standard Fender tone
>> stack with a shorted slope resistor and shorted treble cap, so it wasn't
>> difficult to imagine the result. The killer is the source impedance
>> working against the .033, of course. That swamps everything else,
>> including the effect of the treble control. The source imp. must have
>> been zero'd on the models that look like they don't lose treble, since
>> there's nothing but the little 3.3k resistor to prevent rolloff.
>
>I suspect you're right that I didn't model the source impedance
>correctly. I'll try to recreate the model I made tomorrow and see if
>that makes a difference. I may have changed the values based on my
>reading of this:
>
>http://www.aikenamps.com/ToneControlScaling.html
Randy Aiken is brilliant. Lots of well-presented info on his site.
I suspect that this is not so much a problem with scaling as with
topology. Something is wired 'wrong' (which is to say you failed to
wire in the error in the circuit <g>).
>> Anyway, the curves from the Duncan program appear identical to those
>> from my program. The response is down about -30db at 2k. The mystery is
>> why some report that the circuit works OK. Must be component value
>> errors or problems with the diagram? And the pot still doesn't work as
>> intended on the models, even with lowered source impedance.
>
>Well, I'm convinced. The drawing simply *can't* represent what I put in
>that amp. Either that or my ears are totally out of whack.
Nah, I'm sure your ears are fine. If you have a certain goal in mind,
I'll be glad to work up a circuit for you. I could post it as an
'ascii' schematic, or maybe alter a capture of your original if it
gets more complex. It may actually be better to start from scratch
though, as it's tough to start off by putting a 100k inline with your
signal. The concept of correct impedance scaling requires each
substage to be scaled up by a factor of 3 or 4 as you move left to
right. Otherwise there's too much interaction. IOW, each substage
wants to see a low effective source impedance compared to its own
impedance.
I presume that you're OK with a soldering iron, and that you can get
components, so just post an idea of what you want to accomplish and
I'll get back to you with some ideas for the circuit.
>On Fri, 11 Sep 2009 00:14:24 -0400, RS wrote:
>
>...
>
>> Nah, I'm sure your ears are fine. If you have a certain goal in mind,
>> I'll be glad to work up a circuit for you. I could post it as an
>> 'ascii' schematic, or maybe alter a capture of your original if it gets
>> more complex. It may actually be better to start from scratch though,
>> as it's tough to start off by putting a 100k inline with your signal.
>> The concept of correct impedance scaling requires each substage to be
>> scaled up by a factor of 3 or 4 as you move left to right. Otherwise
>> there's too much interaction. IOW, each substage wants to see a low
>> effective source impedance compared to its own impedance.
>
>Well, I had another go at that circuit simulator and you were dead right.
>With the 100K resistor in place it's flat after about 1k.
>
>Screenshot:
>http://www.flickr.com/photos/22166118@N05/3911182598/sizes/o/
That looks more correct. Still difficult to tell why your amp worked
OK, given that.
>> I presume that you're OK with a soldering iron, and that you can get
>> components, so just post an idea of what you want to accomplish and I'll
>> get back to you with some ideas for the circuit.
>
>Thanks!
>
>The stock-ish VJ sounds a bit boxy at modest volumes, IMO. Not so bad
>when you crank it into overdrive.
>
>I basically want something like the fender tone stack with bass and
>treble fixed at full and a variable mid control but without the overall
>insertion loss. Up to about 20db of cut at a centre frequency switchable
>(or continuously variable) between 400 & 800Hz would be nice. And a
>couple of mojo buttons.
>
>No problem soldering or getting parts. I'll happily try anything you
>cook up even if it involves hand made inductors or whatever. I have a
>new stock VJ combo to experiment on. First though I'm going to have to
>see what's in the the old one and try to figure out why it appears to
>work.
Try this, Yowie. I did this with quick hand-calculations, then
confirmed center frequencies on one of my simulators. I believe it
will do what you want. You'll need to view it in non-proportional
font. Some readers have that option under 'view', or you can capture
it and paste into an editor with non-proportional font. More notes on
the circuit below.
DPDT (linked to section below)
closed = 300hz notch, open = 600hz
| SW1a
C3 120p -----
---||--o o---
C1 | |
0.1uf (DC block) | 120p C2 |
---||---+---------------------+---||--------+---------
| | | |
| | | |
| / R2 / R3 |
| SW1b / 100k / 270k |
| DPDT / / / R4
| C5 | / / / 1 Meg
| .022 ----- | | /<--------
| ---||--+--o o---+ | / Vol
| | / | | /
R1 / | / R5 | | |
250k / | C4 / 2 meg | | |
log /<--+-||---+----------+-------------- |
Mid / .022 |
/ |
| |
| |
--- ---
- -
I took one liberty with your spec: Fender controls center their notch
about 300hz, not 400hz. Also, the 800hz is a bit high, IMO, so I went
with 600hz. So this is still about an octave, at 300hz/600hz. I'll
scale the caps to 400/800 if this doesn't work for you, but I believe
it will. You did say 'Fender', and a lot of that sound comes from the
center frequency that they use.
The notch at 600hz may not always be real useful for clean playing,
but I find that a null in this range can be good when front-ended with
OD pedals to get a more Marshally sound.
If you don't need the 600hz notch, you can use 250p for C2 (omit C3),
and .047 for C4 (omit C5 and R5).
The mid notch will approach -20db on the normal part of its slope, but
phase cancellation at the exact center point will allow you to get a
much deeper notch, in excess of -30db. It's a very useful sound,
especially with a strat in pickup combo positions 2 and 4. It will
smooth out, back up to -20db within the first few degrees of rotation
of the mid pot.
R5 can be any high value. It is there just to keep C5 charged so it
doesn't pop when the switch is pushed. You could use a similar
resistor to keep C3 charged, but it's such a low value that the charge
effect may not be audible. If you model the circuit, you won't need
to add that resistor; it's not doing anything but pop-suppression.
There should be very little insertion loss. I did scale the values in
general Fender range, as you wanted, but with the exception that the
mid control will completely restore mids (Fender's control only has a
few db range, so there's always mid cut). Note that the volume
control is 1 Meg. Not sure why they changed that to 500k in the mod,
as 1 Meg should still have been fine.
The main circuit itself has low component count, but of course the
switched mid-frequency required more parts.
I could work out another variation if you really need continuously
variable mid frequency. It would require a dual pot.
Let me know how it turns out, or how the model looks, if you choose to
do that first.
On 12 Sep 2009 07:08:20 GMT, yowie <yo...@thepub.lan> wrote:
>On Fri, 11 Sep 2009 23:28:40 -0400, RS wrote:
[circuit diagram snipped for brevity. See the "Blackheart Killer Ant
sub 1W tube amp review" thread]
>> Let me know how it turns out, or how the model looks, if you choose to
>> do that first.
>
>It works...
>http://www.flickr.com/photos/22166118@N05/3911802994/sizes/o/
That looks good.
>It may take a bit of hunting around to find a 600V/0.1 cap in Adelaide
>but I can get one online easy enough. Might as well order the other bits
>while I'm at it.
You have some latitude on the value of that cap if you don't mind
losing a db or two of bass. You may not notice much, or may not mind.
In fact, Fender circuits intentionally roll off bass before the signal
gets to the phase splitter. It helps to prevent flubby sounding bass,
and can help more graceful transition into overload.
Try different values for that cap in the emulation and watch what
happens at 82 Hz. That's the fundamental of the Low E on guitar. Keep
in mind that you hear in db, so a linear vertical scale won't really
be representative of the difference in perceived volume.
BTW, you didn't mention the other caps. Do you already have those in
relatively high voltage ratings?
>Thanks again for your trouble. Hopefully I'll build it next weekend.
Sure thing, Yowie. Let me know if it does what you want. If not,
we'll trim the component values.
>On Sat, 12 Sep 2009 04:23:23 -0400, RS wrote:
>
>> Changed the thread title since we've run way off the rails on the
>> Blackheart amp thread hijack.
>
>...
>
>> BTW, you didn't mention the other caps. Do you already have those in
>> relatively high voltage ratings?
>
>I have or can get the other caps rated for 100V (metallised polly). I
>would have thought that was plenty if C1 is doing it's job. Am I missing
>something?
Sorry to say that you need higher voltage caps there, Yowie. I don't
remember the supply voltage but I think it's near 300v. That's
likely to jump if the power tube is pulled (less current through the
dropping resistors). So to be safe, I'd recommend 400v caps (or
higher) all around. And that's presuming the preamp supply is only
300v.
If you were in the US, I'd recommend Digikey or Mouser. I believe
you're in Australia though, right? Not sure who has distribution
there.
Hi Yowie,
That shows 30k input impedance. Did you remove the 100k series
resistor? 30k is more representative of the output impedance of the
tube stage itself, and in fact, may be a bit low at that. With the
100k in series, the total would be 130k+.
Anyway, the difference between the .0033 (3.3nf) and .033 in that spot
is a large one. That may explain why you weren't getting such extreme
high-end rolloff. It would still roll off highs, but over two octaves
up from your previous circuit. Still would be about -20db at 4khz if
the 100k is in series. OK for jazz <g>.
I haven't gone through the details on the site where you've done the
analysis above. What is the yellow trace?
In any event, you'll be better off with a more efficient circuit, and
you'll have more options for adjusting it.
If you're going to get into this for extended experiments, you may
want to keep a few values of high voltage polyester (mylar) caps in
stock. Usually steps of about 1.5x are OK for a start:
.001, .0015, .002, .003, .005, .0068, .0082
.01, .015, .02, .03, .05, (.068, .082 optional)
and some 0.1uf caps for coupling/DC blocking. You already know that
calculated values rarely land right on one of the standard values, so
figure those as approximate. (.002 = .0022, .047 = .05). And there
are component tolerances.
For values smaller than .001, you can use cheap high voltage ceramic
disc caps for a start. I prefer other types, but Fender and others
have used plenty of ceramics over the years.
Resistors are cheap, so you can get whatever you can afford to spend.
In an R-C filter, you can often make up for the wider spread of cap
values by altering an appropriate resistor value, so it's nice to have
5% or 10% jumps within the median range (between, say, 10k through
500k), with some 1M's for doing ground refs (grid to ground, etc).
Pots: 250k, 500k, 1M, mostly audio taper. You may occasionally need a
lower value, but not as often.
If you're dealing with power supply chains, it's nice to have higher
wattage resistors in lower values. Check some schematics for
examples.
Those all relate to tube circuits. Solid state circuits would
typically be lower in value and cover a wider range.
Of course I should have posted this before you placed your order, eh?
When I saw "Made in Korea" on my head and cab, I assumed they were all
made in Korea. Not true. Some are made in China. I don't know dates
or details on when a switch was made. But my head had a production date
of Jan. 09.
>On Tue, 15 Sep 2009 16:26:35 -0400, RS wrote:
>
>> On 15 Sep 2009 10:36:51 GMT, yowie <yo...@thepub.lan> wrote:
>>>
>>>Oh. And I think I must have built the original circuit like this:
>>>
>>>http://www.flickr.com/photos/22166118@N05/3921990197/sizes/o/
>>>
>>>...with both caps 3.3n. Will find out on Saturday.
>>
>> Hi Yowie,
>>
>> That shows 30k input impedance. Did you remove the 100k series resistor?
>
>No. I was in a hurry and forgot to put it in.
Ah. BTW, the typical output impedance for a 12ax7 with 100k load
resistor should be around 38k or so, if memory serves.
>I just had another look at the thread on sewatt.com. A couple of folks
>reported building the circuit as originally shown and that it didn't
>work. Further down someone else provides the following advice:
>
>===============
>http://www.sewatt.com/files/sewatt/Mid%20Notch%20pot%20schem.png
>...
>To work out the parameters for this filter, use the following formular
>(sic)
[stuff snipped for now]
>... without noting that the original drawing showed 'C' as 2 different
>values. Any comments?
Tough to comment, given the problems with the circuit. It's basically
a "bridged T" filter, but -with all the components backward!- IOW,
the resistors are caps and vice versa. The circuit, as given, would
not scale well to even low source impedances. The conventional
bridged T would work fine. So I'm not sure why they did that.
The math for the regular bridged T would apply to the filter that I
posted for you. If you're interested, I could explain that. It's
straightforward ratiometric stuff for computing the center frequency,
since the values are scaled for minimal interaction between the R-C
and C-R components. The response can also be approximated by simple
ratios, too, except for the deep notch exactly at center, which is
caused by phase shift cancellation.
>> In any event, you'll be better off with a more efficient circuit, and
>> you'll have more options for adjusting it.
>
>... [noted]
>
>> Of course I should have posted this before you placed your order, eh?
>
>I ordered enough caps to build two. If it works out I'll probably just
>add switchable NFB and get on with actually playing (the guitar) for a
>while :).
Sounds like a good idea. Please get back when you get the new circuit
wired in, OK?
I have one of the earliest versions. Made in China with Chinese tubes.
Great build quality and sound. I would prefer a little less muddiness and
a bit more high end, but still it's OK.
I wonder how the two versions compare. Korea or China.
I swapped out the tubes for Tungsols, and I think it sound a bit better,
but the original tubes weren't too bad.
I have some Mullard, JJ, and various other tubes in varioues 12A_7
variation that I will try, but I already enjoy playing it with the
Tungsols, and, seriously, who has the time ...
It is now my main amp because I don't want to go deaf playing a 100watt
full stack to get "the tone" and I don't want the neighbors hating "That
guy" who is disturbing the peace.
Yeah, the Blackheart Little Giant was too loud. The kids two blocks away
said it was too loud, even!
Highly recommended to keep the peace.
Epiphone Valve Junior -Too loud
Blackheart Little giant too loud
Fender Champ Ok but speaker sucks.
Crate V5 - just sucks.
I do love my new VOX AC4TV. It sounds brighter and less muddy, but less
low end. I can set it anywhere from 4watts down to .25 watts, same as the
Killer ant, but the VOX does sounds muddy at the lowest level.
There is an 8" speaker model that can go down to 0.1 watt, but the 8"
speaker doesn't sound as good.
hah, who are you kidding! first upgrade the caps and resistors, speaker
upgrade, new tubes. . .
>On Wed, 16 Sep 2009 19:02:46 -0400, RS wrote:
>
>> On 15 Sep 2009 22:02:30 GMT, yowie <yo...@thepub.lan> wrote:
>
>...
>
>>>I ordered enough caps to build two. If it works out I'll probably just
>>>add switchable NFB and get on with actually playing (the guitar) for a
>>>while :).
>>
>> Sounds like a good idea. Please get back when you get the new circuit
>> wired in, OK?
>
>Done. I finally got around to building it last night. No surprise, it
>works fine. I haven't had the chance to turn it up much as yet but at
>low volumes it's great. There is about the same loss in the circuit as
>there was with the R7/R7 voltage divider that was in the same place -
>maybe a tad more.
Should be less overall loss when the mid control is at max.
>Instead of that one nasal honk it goes from 'warm' to
>alarmingly 'funky' with usable tones in between.
Very good then.
> This is with just the
>300Hz centre - haven't put in the switch yet since I'm reluctant to drill
>holes until I'm sure I'll want them.
I think the 300hz center will be of more use to you than the 600hz
position. But both should be useful. As mentioned, the 600hz notch
may be good for sort of a Marshall-like phasey overdrive. Probably
not as much use for clean. But as always, it's subjective.
>Thanks again for your efforts. Do you mind if I pass this design along
>to the folks at sewatt.com? They have a persistent thread there devoted
>to mid cut circuit ideas.
Sure, Yowie. Go for it. Glad that it worked for you.