Of course you can go much farther if you are a FAB*.
I have a nice F7 here with -9 damping factor.
* Fearless Amplifier Builder
Mr Pass, please feel free to email (pm) me. I have a chassis and power supply ready. Would just have to p2p the boards. 😀
This is DIY. 
I'll tell what what. Attached is a simple schematic, showing an arbitrarily
simple gain stage.
The damping factor is determined by a confluence of values. LHQUAM
has put together a nice thread discussing this in mathematical detail -
I refer you to his posts.
R3/R4 sets the gain of the "op amp", but it could be a stage with fixed
gain without reference to R3/R4. It could be any sort of non-inverting
power amplifier with reasonable gain, bandwidth and damping factor.
Damping factor is generally influenced by the following:
An increase in the fixed gain of the amplifier will increase the DF
An increase in R1/R2 will increase the DF. Remember that R1 includes
the signal source impedance.
An increase in R5/Rload (speaker) will increase the DF.
As you increase these, the output impedance seen by the load
will down until it literally reaches 0, at which the DF becomes infinity.
More will take you into the region of negative output impedance where
you have negative damping factors, starting with -infinity and proceeding
to "lower" negative DF values.
When you get into the negative region, things get different.
As you go into the negative output impedances the amplifier starts to
exhibit motional control over the voice coil. A DF of -5 has more of this
control than -10, for example.
Low damping factors (like current source amps) encourage acceleration
control. Negative damping factors move you toward velocity control.
So go forth, plug this schematic into your simulators.
Or if you are a paid up member of FAB, wire it up and remember not to
inhale the smoke.
Whoops, I have a reservation at Trink's.
I'll tell what what. Attached is a simple schematic, showing an arbitrarily
simple gain stage.
The damping factor is determined by a confluence of values. LHQUAM
has put together a nice thread discussing this in mathematical detail -
I refer you to his posts.
R3/R4 sets the gain of the "op amp", but it could be a stage with fixed
gain without reference to R3/R4. It could be any sort of non-inverting
power amplifier with reasonable gain, bandwidth and damping factor.
Damping factor is generally influenced by the following:
An increase in the fixed gain of the amplifier will increase the DF
An increase in R1/R2 will increase the DF. Remember that R1 includes
the signal source impedance.
An increase in R5/Rload (speaker) will increase the DF.
As you increase these, the output impedance seen by the load
will down until it literally reaches 0, at which the DF becomes infinity.
More will take you into the region of negative output impedance where
you have negative damping factors, starting with -infinity and proceeding
to "lower" negative DF values.
When you get into the negative region, things get different.
As you go into the negative output impedances the amplifier starts to
exhibit motional control over the voice coil. A DF of -5 has more of this
control than -10, for example.
Low damping factors (like current source amps) encourage acceleration
control. Negative damping factors move you toward velocity control.
So go forth, plug this schematic into your simulators.
Or if you are a paid up member of FAB, wire it up and remember not to
inhale the smoke.
Whoops, I have a reservation at Trink's.
Interesting Mr Pass and bonne appétit at Trink's !
Do i take risk if huge damping factor velocity force shake drivers too much ? A joke
That is indeed correct. I checked their website in the meantime. For the nights they serve dinner you need a reservation for parties of six or more.
When I go it's usually just for lunch. Their desserts and coffee are both top notch.
Graeme
When I go it's usually just for lunch. Their desserts and coffee are both top notch.
Graeme
Like 1812 Overture trick- pause the cd just before the canon shot, turn volume to full on 300w amp, then hit play.
I am using positive current feedback [PCF] in a diy Class aP [analog Pulse] amp. The Type 1 scheme posted by lhquam in the thread "First Watt F7 review" was used therein. Will post a blurb in its thread.
The effect of this PCF-enhanced speaker damping on fidelity [clarity, resolution] is truly astounding.
Use PCF in Class D too!
Need to protect the Gate-Source Junction of SJDP120R085
I damaged an SJDP120R085 in a DEF amp during the practice of positive current feedback [PCF] as I had described in earlier posts. The DEF amp was idling at the standard conditions [+/1.3A Id at +/-23Vds] when I made a change [I thought was subtle] which caused it to oscillate driving a loudspeaker.
The original idle Id current [1.3A] dropped to ~50 mA and the output voltage offset rose to >50 mV. The two 2 A fast rail fuses didn't blow.
1. The damaged R085 had developed a substantial gate current.
2. The original value of Vds for each FET was -3.7V [PMOS] and -4.34V [R085]
3. The post damage Vds for PMOS =-3.2V and -3.7V for R085 while using the same resistor string to generate the bias. . This resistor string could no longer sustain the parent Vds values because its flowing current was being drained in a new direction.
4. PMOS was not damaged. A Vds =-3.6 V from a pre-charged 220uF stayed put with time.
5. An initial Vds for R085 = -3.7V from a second pre-charged 220uf began to visibly decrease with time.
A standard 6.2V Zener will protect the Gate -Source junction of R085. It is higher in value than its Pinch-off of -5Vdc. The positive end of the Zener goes to the Source and its negative end goes to the Gate of R085. Recall, that a 100 Ohm resistor is in series with the gate of R085 and will protect the Zener from overcurrent.
I damaged an SJDP120R085 in a DEF amp during the practice of positive current feedback [PCF] as I had described in earlier posts. The DEF amp was idling at the standard conditions [+/1.3A Id at +/-23Vds] when I made a change [I thought was subtle] which caused it to oscillate driving a loudspeaker.
The original idle Id current [1.3A] dropped to ~50 mA and the output voltage offset rose to >50 mV. The two 2 A fast rail fuses didn't blow.
1. The damaged R085 had developed a substantial gate current.
2. The original value of Vds for each FET was -3.7V [PMOS] and -4.34V [R085]
3. The post damage Vds for PMOS =-3.2V and -3.7V for R085 while using the same resistor string to generate the bias. . This resistor string could no longer sustain the parent Vds values because its flowing current was being drained in a new direction.
4. PMOS was not damaged. A Vds =-3.6 V from a pre-charged 220uF stayed put with time.
5. An initial Vds for R085 = -3.7V from a second pre-charged 220uf began to visibly decrease with time.
A standard 6.2V Zener will protect the Gate -Source junction of R085. It is higher in value than its Pinch-off of -5Vdc. The positive end of the Zener goes to the Source and its negative end goes to the Gate of R085. Recall, that a 100 Ohm resistor is in series with the gate of R085 and will protect the Zener from overcurrent.
@ Antoinel
Semisouth are precious thanks for share your experiance.
That be never happen again in the future
🙂
Semisouth are precious thanks for share your experiance.
That be never happen again in the future
🙂
A zener is standard for the Jfets, even though you are not dealing with the
thin oxide Gate of Mosfets. I have been able to blow out SemiSouth Jfets
and SITs by applying reverse G/S voltage above 40 volts or so.
I lost a couple SIT amps from this early on before I came to realize that there
are actually tube preamps out there that deliver that sort of turn-on/turn-off
transient. A 1 watt 9.1V Zener took care of that, and has been standard for
power Jfets since.
