BAJA Ampeg SVT bass amplifier pedal 9v 010418 [documentation]
- bajaman
- Old Solderhand
Information
- Posts: 4549
- Joined: 26 Jun 2007, 21:18
- Location: New Brighton, Christchurch, NZ
- Has thanked: 595 times
- Been thanked: 2058 times
Here is the second installment of my bass amplifier series - the Ampeg SVT in 9v pedal friendly format.
Plug this pedal into any high wattage power amplifier and experience the Super Valve Technology sound without tubes
This design is based on the original Ampeg SVT but with the 5 frequency bands mid boost / cut rotary switch used in the later SLM MkII model.
I left the bass cut option in the ultra bass switch control out as well just as they did in the MkII However I modeled the gain and phase relationships of the original amplifier which used the 12WD7 tubes as much as possible. One thing I feel worth mentioning is the back to back clipping diodes used by Ampeg at the output of the preamplifier section. these diodes are no longer easily available but have very similar characteristics to the good old standard 1N4148 devices, and these are what i have used in the pedal version. They have a Vf of 1 volt according to the datasheets but in practice I have found them to conduct at slightly lower voltage. A very interesting thing with the Ampeg design is that these diodes clamp the preamp signal voltage just before the output tubes and so the clipping is predominantly due to these components NOT the output power amplifier section I do not recall ever seeing this used in any other tube bass head by other manufacturers. The Ampeg SVT power amplifier stage has about 43dB voltage gain and apart from a slight high and low end rolloff accurately and transparently tracks the preamplifier response and phase characteristic. For this reason I have made no attempt to model the power amplifier section behaviour in this pedal other than duplicating the low end and high end rolloffs in the final buffer stage of this design.
Another thing worth mentioning In the Ampeg SVT, there is a cathode follower buffer stage following the mid range section. In my design I have used a gain stage in this position to raise the signal level enough to clip the back to back diodes at exactly the same level as in the original tube amplifier I had to reduce the gain of the three preceeding stages to avoid clipping the op amps and this is why the need for the additional gain stage after the tone controls. Working with 9 volts instead of 300 volts means careful consideration of gain structure is paramount in a design such as this.
Another interesting observation - the bright input actualy reduces the low end response of the SVT, and this gives the impression of a brighter sounding amplifier . instead of having two input sockets, I decided to use a single input socket and incorporate a bright / normal mini toggle switch
I changed the mid frequency boost / cut frequencies slightly - 220Hz, 440Hz, 880Hz, 1k8Hz and 3k5Hz - as close as possible to octaves of "A 440 "
Enough waffling - here it is - i will build and verify this in the next few weeks - waiting on some inductors to arrive from China ( I intend to use five of these 10mH coils in a series chain https://www.ebay.com/itm/10pcs-Inductor ... 2749.l2649)
cheers
bajaman
Plug this pedal into any high wattage power amplifier and experience the Super Valve Technology sound without tubes
This design is based on the original Ampeg SVT but with the 5 frequency bands mid boost / cut rotary switch used in the later SLM MkII model.
I left the bass cut option in the ultra bass switch control out as well just as they did in the MkII However I modeled the gain and phase relationships of the original amplifier which used the 12WD7 tubes as much as possible. One thing I feel worth mentioning is the back to back clipping diodes used by Ampeg at the output of the preamplifier section. these diodes are no longer easily available but have very similar characteristics to the good old standard 1N4148 devices, and these are what i have used in the pedal version. They have a Vf of 1 volt according to the datasheets but in practice I have found them to conduct at slightly lower voltage. A very interesting thing with the Ampeg design is that these diodes clamp the preamp signal voltage just before the output tubes and so the clipping is predominantly due to these components NOT the output power amplifier section I do not recall ever seeing this used in any other tube bass head by other manufacturers. The Ampeg SVT power amplifier stage has about 43dB voltage gain and apart from a slight high and low end rolloff accurately and transparently tracks the preamplifier response and phase characteristic. For this reason I have made no attempt to model the power amplifier section behaviour in this pedal other than duplicating the low end and high end rolloffs in the final buffer stage of this design.
Another thing worth mentioning In the Ampeg SVT, there is a cathode follower buffer stage following the mid range section. In my design I have used a gain stage in this position to raise the signal level enough to clip the back to back diodes at exactly the same level as in the original tube amplifier I had to reduce the gain of the three preceeding stages to avoid clipping the op amps and this is why the need for the additional gain stage after the tone controls. Working with 9 volts instead of 300 volts means careful consideration of gain structure is paramount in a design such as this.
Another interesting observation - the bright input actualy reduces the low end response of the SVT, and this gives the impression of a brighter sounding amplifier . instead of having two input sockets, I decided to use a single input socket and incorporate a bright / normal mini toggle switch
I changed the mid frequency boost / cut frequencies slightly - 220Hz, 440Hz, 880Hz, 1k8Hz and 3k5Hz - as close as possible to octaves of "A 440 "
Enough waffling - here it is - i will build and verify this in the next few weeks - waiting on some inductors to arrive from China ( I intend to use five of these 10mH coils in a series chain https://www.ebay.com/itm/10pcs-Inductor ... 2749.l2649)
cheers
bajaman
be kind to all animals - especially human beings
- bajaman
- Old Solderhand
Information
- Posts: 4549
- Joined: 26 Jun 2007, 21:18
- Location: New Brighton, Christchurch, NZ
- Has thanked: 595 times
- Been thanked: 2058 times
Hi folks
built now and verified - sounds just like all the SVT amps i have worked on too
A few mistakes and modifications
first up - I left an 820 ohm resistor off the vero layout around op amp 3b, so here is the amended vero second - I found the treble control to be very ineffective - not sure why so i decided to implement the SVT MkII treble circuit capacitors and got a much more useable treble control, so here is the updated rev 1 schematic and the updated parts not on vero finally here are some build pictures - I decided to wind the mid range inductor - I had some Amidon pot cores and 34 gauge enamelled copper wire lying around - it only took a few minutes to hand wind.
winding details are: Amidon Associates Inc. #PC-1811-77 (ferrite pot core - material #77, 2000 permeability)
53 turns for first tap at 10mH, a further 38 turns for the second tap at 30mH and a further 26 turns for the end wind to give 50mH in total.
34 (or similar) gauge enamelled copper wire
The first picture shows the pot core mounted on some 3M double sided foam tape in the middle of the top panel On my build i decided to make the bright / normal input switch footswitchable using a 3Pdt footswitch - I added an additional 10k resistor on the third section of the footswitch around the feedback loop of the first op amp to increase the gain to compensate for the loss of low end when the bright input is engaged. Hope that makes sense ( 4k7 for normal and 10k plus 4k7 in series for bright) - the footswitch simply shorts the 10k resistor when normal is selected. This 10k resistor could be tweaked and allows a gain ( and slight treble) boost for function for soloing happy building
cheers
bajaman
built now and verified - sounds just like all the SVT amps i have worked on too
A few mistakes and modifications
first up - I left an 820 ohm resistor off the vero layout around op amp 3b, so here is the amended vero second - I found the treble control to be very ineffective - not sure why so i decided to implement the SVT MkII treble circuit capacitors and got a much more useable treble control, so here is the updated rev 1 schematic and the updated parts not on vero finally here are some build pictures - I decided to wind the mid range inductor - I had some Amidon pot cores and 34 gauge enamelled copper wire lying around - it only took a few minutes to hand wind.
winding details are: Amidon Associates Inc. #PC-1811-77 (ferrite pot core - material #77, 2000 permeability)
53 turns for first tap at 10mH, a further 38 turns for the second tap at 30mH and a further 26 turns for the end wind to give 50mH in total.
34 (or similar) gauge enamelled copper wire
The first picture shows the pot core mounted on some 3M double sided foam tape in the middle of the top panel On my build i decided to make the bright / normal input switch footswitchable using a 3Pdt footswitch - I added an additional 10k resistor on the third section of the footswitch around the feedback loop of the first op amp to increase the gain to compensate for the loss of low end when the bright input is engaged. Hope that makes sense ( 4k7 for normal and 10k plus 4k7 in series for bright) - the footswitch simply shorts the 10k resistor when normal is selected. This 10k resistor could be tweaked and allows a gain ( and slight treble) boost for function for soloing happy building
cheers
bajaman
be kind to all animals - especially human beings
- nickthemen
- Breadboard Brother
Nice! Could u explain the copper shielding, and is that not making a groundloop?
Gr
Gr
- bajaman
- Old Solderhand
Information
- Posts: 4549
- Joined: 26 Jun 2007, 21:18
- Location: New Brighton, Christchurch, NZ
- Has thanked: 595 times
- Been thanked: 2058 times
only there to establish a ground connection to the enclosure for shielding purposes - if i had used metal jack sockets it would have not been necessaryCould u explain the copper shielding
be kind to all animals - especially human beings
- bajaman
- Old Solderhand
Information
- Posts: 4549
- Joined: 26 Jun 2007, 21:18
- Location: New Brighton, Christchurch, NZ
- Has thanked: 595 times
- Been thanked: 2058 times
entirely up to youbucksears wrote:Is this strictly to be used with a power amp, or can it be used as a dirt pedal / AIAB in front of an existing bass amp?
cheers
bajaman
be kind to all animals - especially human beings
Hi Baja,
A great one, thanks a lot! Some questions though... Is there a reason you made the input buffer out of a jfet instead of using IC1b for this? Also I'm thinking about making this one the preamp of a Class D poweramp I've seen on ebay. I would use a switching powersupply for it (again from the evilbay) and those have +/-15V outputs too. Do you think I could adapt this one to bipolar powersupply? I think I would only need to remove the vref connections from the inputs of the opamps and change all other vref connections to ground instead. And I wouldn't need a vref at all in that case I guess... Or am I wrong here? Thanks!
A great one, thanks a lot! Some questions though... Is there a reason you made the input buffer out of a jfet instead of using IC1b for this? Also I'm thinking about making this one the preamp of a Class D poweramp I've seen on ebay. I would use a switching powersupply for it (again from the evilbay) and those have +/-15V outputs too. Do you think I could adapt this one to bipolar powersupply? I think I would only need to remove the vref connections from the inputs of the opamps and change all other vref connections to ground instead. And I wouldn't need a vref at all in that case I guess... Or am I wrong here? Thanks!
Thank you, Bajaman! Very nice of you to include schematic!
Also, seconded for wondering about +/-15vdc for more headroom and clarity. Would love to add this charge pump that Ive used on other preamps:
http://tagboardeffects.blogspot.com/201 ... erter.html
Also, seconded for wondering about +/-15vdc for more headroom and clarity. Would love to add this charge pump that Ive used on other preamps:
http://tagboardeffects.blogspot.com/201 ... erter.html
Baja!
Thanks for putting together a neat and VERY interesting build!
I've been using the SCR-DI for quite some time. While it is good, it lacks the midrange selector of the SVT, as well as the ability to get any sort of real "grunt" out of it (save for the Scrambler, which isn't very good by itself and should never have been given THAT name). It DOES add a bit of that old "pushed to the edge" sound... But it is only an approximation, and MUST be used with the blend control ( I have to run Drive and Blend at noon to get anywhere near what I hear as use-able)!
That being said, I look forward to start "kitting this out" as an alternative.
Judging from what I've read over at Tagboard, you've dotted your I"i's" and crossed the "t's" on this!
BTW, anyone got a source on 20uH through-hole inductors. Digi-Key has 10uH ones, but nothing more (at least that I could find)?
Thanks for putting together a neat and VERY interesting build!
I've been using the SCR-DI for quite some time. While it is good, it lacks the midrange selector of the SVT, as well as the ability to get any sort of real "grunt" out of it (save for the Scrambler, which isn't very good by itself and should never have been given THAT name). It DOES add a bit of that old "pushed to the edge" sound... But it is only an approximation, and MUST be used with the blend control ( I have to run Drive and Blend at noon to get anywhere near what I hear as use-able)!
That being said, I look forward to start "kitting this out" as an alternative.
Judging from what I've read over at Tagboard, you've dotted your I"i's" and crossed the "t's" on this!
BTW, anyone got a source on 20uH through-hole inductors. Digi-Key has 10uH ones, but nothing more (at least that I could find)?
- bajaman
- Old Solderhand
Information
- Posts: 4549
- Joined: 26 Jun 2007, 21:18
- Location: New Brighton, Christchurch, NZ
- Has thanked: 595 times
- Been thanked: 2058 times
You need 10mH and 20mH (not uH) - winding details for the inductor are earlier in the thread or you could get 10mH inductors easily on Aliexpressbasscapes wrote:Baja!
Thanks for putting together a neat and VERY interesting build!
I've been using the SCR-DI for quite some time. While it is good, it lacks the midrange selector of the SVT, as well as the ability to get any sort of real "grunt" out of it (save for the Scrambler, which isn't very good by itself and should never have been given THAT name). It DOES add a bit of that old "pushed to the edge" sound... But it is only an approximation, and MUST be used with the blend control ( I have to run Drive and Blend at noon to get anywhere near what I hear as use-able)!
That being said, I look forward to start "kitting this out" as an alternative.
Judging from what I've read over at Tagboard, you've dotted your I"i's" and crossed the "t's" on this!
BTW, anyone got a source on 20uH through-hole inductors. Digi-Key has 10uH ones, but nothing more (at least that I could find)?
be kind to all animals - especially human beings
Information
Built and used this at a rehearsal last night and really pleased with it. Used it in the effect return of my bass amp bypassing the pre-amp and all sounds very nice.
I have noticed that the mid switching using the inductors has a very subtle effect, hardly noticable, Is this correct? Also the high and low switches there is no noticeable difference in sound. The bright switch definitely has a big change in the sound. Just wanted to make sure this sounds about right and there is nothing wrong with my build.
Best thing is it has "that sound"
I have noticed that the mid switching using the inductors has a very subtle effect, hardly noticable, Is this correct? Also the high and low switches there is no noticeable difference in sound. The bright switch definitely has a big change in the sound. Just wanted to make sure this sounds about right and there is nothing wrong with my build.
Best thing is it has "that sound"
Information
In reply to myself looks like I missed the return wire from the rotary selector switch back to the main board! fix this later
- temol
- Solder Soldier
I have it on a breadboard now but there is some kind of clipping/distortion/excitation issue. I've breadboarded this circuit three times and this effect is present every time. It's like high frequency distortion or "too much treble" effect. It dissapears when I remove 10n from upper treble branch. And it propagates all the way, to the IC1A. Any ideas?
T.
T.
By the way, the question is really interesting. I also wonder why not use IC1b for the buffer.drd wrote:Hi Baja,
A great one, thanks a lot! Some questions though... Is there a reason you made the input buffer out of a jfet instead of using IC1b for this? Also I'm thinking about making this one the preamp of a Class D poweramp I've seen on ebay. I would use a switching powersupply for it (again from the evilbay) and those have +/-15V outputs too. Do you think I could adapt this one to bipolar powersupply? I think I would only need to remove the vref connections from the inputs of the opamps and change all other vref connections to ground instead. And I wouldn't need a vref at all in that case I guess... Or am I wrong here? Thanks!
Very interesting to know the thoughts of the author.
- bajaman
- Old Solderhand
Information
- Posts: 4549
- Joined: 26 Jun 2007, 21:18
- Location: New Brighton, Christchurch, NZ
- Has thanked: 595 times
- Been thanked: 2058 times
Simple answer: because IC1b is already being used as a low impedance vbias source. However most any small signal njfet could be used as the input buffer, or you could simply omit using the buffer at all if you like. The 2SK117 100k source resistor could be reduced as low as 10k if desired. If using a bipolar power supply (+/- 15v), yes your assumption is correct.zyu wrote:By the way, the question is really interesting. I also wonder why not use IC1b for the buffer.drd wrote:Hi Baja,
A great one, thanks a lot! Some questions though... Is there a reason you made the input buffer out of a jfet instead of using IC1b for this? Also I'm thinking about making this one the preamp of a Class D poweramp I've seen on ebay. I would use a switching powersupply for it (again from the evilbay) and those have +/-15V outputs too. Do you think I could adapt this one to bipolar powersupply? I think I would only need to remove the vref connections from the inputs of the opamps and change all other vref connections to ground instead. And I wouldn't need a vref at all in that case I guess... Or am I wrong here? Thanks!
Very interesting to know the thoughts of the author.
cheers
bajaman
be kind to all animals - especially human beings
- jhergonz
- Breadboard Brother
I think this is a perfect timing to ask this... What is the advantage of having a buffer in vref? Or low impedance vref?bajaman wrote:Simple answer: because IC1b is already being used as a low impedance vbias source. However most any small signal njfet could be used as the input buffer, or you could simply omit using the buffer at all if you like. The 2SK117 100k source resistor could be reduced as low as 10k if desired. If using a bipolar power supply (+/- 15v), yes your assumption is correct.zyu wrote:By the way, the question is really interesting. I also wonder why not use IC1b for the buffer.drd wrote:Hi Baja,
A great one, thanks a lot! Some questions though... Is there a reason you made the input buffer out of a jfet instead of using IC1b for this? Also I'm thinking about making this one the preamp of a Class D poweramp I've seen on ebay. I would use a switching powersupply for it (again from the evilbay) and those have +/-15V outputs too. Do you think I could adapt this one to bipolar powersupply? I think I would only need to remove the vref connections from the inputs of the opamps and change all other vref connections to ground instead. And I wouldn't need a vref at all in that case I guess... Or am I wrong here? Thanks!
Very interesting to know the thoughts of the author.
cheers
bajaman