Re: kt 88 schematics
I think all the answers that you need are in this thread. If you are in fact looking for 'tried and true', build up a Dynaco Mark 3.
The only Hammond power trannies that would work is the 278x if you are running a tube rectifier, or the 278cx if you are not.
Triode Electronics (triodeelectronics.com) has everything that you need for a Dynaco... right down to pre-punched chassis, boards and even mods and upgrades. They have the original reproduction trannys, but you seem set on Hammond, so use those instead.
The reason that you don't see classic designs with Hammond is that there are much better trannys out there. Hammond are just cheap and available and they have a wide product line. You are mostly going to find experimental amps and DIY one-offs built with Hammond. (NOTE: I am not knocking them... I have literally dozens of Hammonds sitting around me as i type this)
jbldriven said:
I think all the answers that you need are in this thread. If you are in fact looking for 'tried and true', build up a Dynaco Mark 3.
The only Hammond power trannies that would work is the 278x if you are running a tube rectifier, or the 278cx if you are not.
Triode Electronics (triodeelectronics.com) has everything that you need for a Dynaco... right down to pre-punched chassis, boards and even mods and upgrades. They have the original reproduction trannys, but you seem set on Hammond, so use those instead.
The reason that you don't see classic designs with Hammond is that there are much better trannys out there. Hammond are just cheap and available and they have a wide product line. You are mostly going to find experimental amps and DIY one-offs built with Hammond. (NOTE: I am not knocking them... I have literally dozens of Hammonds sitting around me as i type this)
Thanks for your help. I am not stuck on the dynaco circuit and willing to assemble even build my own chassi. I look at this as therapy from my day job and is probaly cheaper then proffessional help. There is something euphoric when the solder is burning and the pieces are coming together.
I will reseach the site you provided
I also order the book from London Power to review circuit.
IF anyone else can provide additional informs please do
Thanks again
I will reseach the site you provided
I also order the book from London Power to review circuit.
IF anyone else can provide additional informs please do
Thanks again
I built 2 monoblocks based on Leak TL25+ (EL34) using Hammond 1650H output and 374BX mains tranny.
In due course I've found that I had to adjust the original NFB values to obtain a more natural sound. Without adjusting the original values, the amp did sound rather aggressive, fatiguing, vocals seem shouty and sound of certain music instruments being rough or edgy to the point of sounding unnatural. In short, it just wasn't sweet enough although I though it's a damn good 30+ watter amp! I felt it could be better by tweaking some component values before I start thinking about better output trannys and whether I'm going to reap benefits in due course.
I'm convinced any builder shouldn't take the original values for granted, but as a starting point for further improvement. The cost those resistors and silver mica caps cost a fraction of an new quality tranny.
For now I don't have an o'scope to check the sine/square waveform but will do so very soon. I've spent lots of time trial & error by ear and it sounds better by a huge, HUGE margin. I think I'm quite happy with the results I've gotten. Having said that, I now wonder how much more good can it get IF with better quality transformers and going through the vicious cycle of tweaking it all over again..I think my hair will turn white by then and a very thin bank balance.
I think anyone DIY'ing a tube amp with aftermarket parts/output trafo, cloning a tried & tested circuit must look into altering the NFB values (especially the resistor bypass cap) if neccessary.
In due course I've found that I had to adjust the original NFB values to obtain a more natural sound. Without adjusting the original values, the amp did sound rather aggressive, fatiguing, vocals seem shouty and sound of certain music instruments being rough or edgy to the point of sounding unnatural. In short, it just wasn't sweet enough although I though it's a damn good 30+ watter amp! I felt it could be better by tweaking some component values before I start thinking about better output trannys and whether I'm going to reap benefits in due course.
I'm convinced any builder shouldn't take the original values for granted, but as a starting point for further improvement. The cost those resistors and silver mica caps cost a fraction of an new quality tranny.
For now I don't have an o'scope to check the sine/square waveform but will do so very soon. I've spent lots of time trial & error by ear and it sounds better by a huge, HUGE margin. I think I'm quite happy with the results I've gotten. Having said that, I now wonder how much more good can it get IF with better quality transformers and going through the vicious cycle of tweaking it all over again..I think my hair will turn white by then and a very thin bank balance.
I think anyone DIY'ing a tube amp with aftermarket parts/output trafo, cloning a tried & tested circuit must look into altering the NFB values (especially the resistor bypass cap) if neccessary.
Thanks for the information. I still hope to find someone out there with experience in building KT88 Amps with Hammond tranys..
I'm new what does NFB mean ?
I have 1650 p outputs, kt88 tubes, sockets and 278cx primary. However i don't have the experience to design the circuits.So i need expert help. With a proven circuit and clear parts identifications I am capable of pulling this off.
There seems to be alot of experience on this site yet no one has built this combo.
I am waiting to see the Principals of Power Book for the schematics they have. Thanks to GINGERTUBE I may be heading down the right path.
If any one has any other ideas or help please step in
I'm new what does NFB mean ?
I have 1650 p outputs, kt88 tubes, sockets and 278cx primary. However i don't have the experience to design the circuits.So i need expert help. With a proven circuit and clear parts identifications I am capable of pulling this off.
There seems to be alot of experience on this site yet no one has built this combo.
I am waiting to see the Principals of Power Book for the schematics they have. Thanks to GINGERTUBE I may be heading down the right path.
If any one has any other ideas or help please step in
jbldriven,
Go to www.diytube.com
Click on the eiclone thread.
In that area you'll find two threads on Ikezilla.
Uses PPP KT-88's for about 120 watts out with a Hammond OPT.
The output stage circuitry can be found in those threads. The driver circuit can be found with the original eiclone documents at the top of the page.
Hope this helps.
Regards,
Ed Long
Go to www.diytube.com
Click on the eiclone thread.
In that area you'll find two threads on Ikezilla.
Uses PPP KT-88's for about 120 watts out with a Hammond OPT.
The output stage circuitry can be found in those threads. The driver circuit can be found with the original eiclone documents at the top of the page.
Hope this helps.
Regards,
Ed Long
If you have little/no experience, I'd recommend doing this project in phases.
For Phase 1, I'd do KT88's stapped as triode and cathode biased. This will
get you started with 20-30W. The advantage is that there is little that can go wrong:
Cathode Bias will be more forgiving to mistakes that might otherwise put your
KT88s or Transformer at risk. With triode mode, you likely wont need any feedback and you can get your chops up with adding feedback. This will also
give you the chance to get your power supply right.
The more power you shoot for the more likelihood that power will break something
if you do it wrong. So, I'd advise starting triode mode, cathode biased. From there,
you can upgrade to Ultralinear with NFB with no lost parts and some small changes.
How about: KT88 cathode biased with 47 ohm 2W resistor between plate and screen grid. For a driver, 12BH7/6GU7 or 5687. God knows what for a phase splitter. If this amp is being driven by Pro audio gear (i.e. balanced line out), then
you could use an Edcor WSM600/10K transformer for both balanced input and phase splitter.
-- Jim
For Phase 1, I'd do KT88's stapped as triode and cathode biased. This will
get you started with 20-30W. The advantage is that there is little that can go wrong:
Cathode Bias will be more forgiving to mistakes that might otherwise put your
KT88s or Transformer at risk. With triode mode, you likely wont need any feedback and you can get your chops up with adding feedback. This will also
give you the chance to get your power supply right.
The more power you shoot for the more likelihood that power will break something
if you do it wrong. So, I'd advise starting triode mode, cathode biased. From there,
you can upgrade to Ultralinear with NFB with no lost parts and some small changes.
How about: KT88 cathode biased with 47 ohm 2W resistor between plate and screen grid. For a driver, 12BH7/6GU7 or 5687. God knows what for a phase splitter. If this amp is being driven by Pro audio gear (i.e. balanced line out), then
you could use an Edcor WSM600/10K transformer for both balanced input and phase splitter.
-- Jim
KT88 Hammond Questions
Getting closer to starting and need other simple questions answered possibly stupid ones to.
1. Is the pin number on a kt 88 start at @#1 clockwise from the alignment pin in the center. Is the same for all tubes.
2. How do you determine best types and brands for the capacitors in the circuits. and where they sould be used. ie power supply, bias circuit, etc
3. How does this also work with resistors.
4. Do i use 105 degree C, solid or stranded wire, 18 or 20gauge 600volt.
6.I want to use eyelet circuit board system, The standard thickness is 1/16" Where can i buy the eyelets and flaring tool. I would like to possibly use 1/8" material is there eyelets available.
I. know where to buy the board material
Getting closer to starting and need other simple questions answered possibly stupid ones to.
1. Is the pin number on a kt 88 start at @#1 clockwise from the alignment pin in the center. Is the same for all tubes.
2. How do you determine best types and brands for the capacitors in the circuits. and where they sould be used. ie power supply, bias circuit, etc
3. How does this also work with resistors.
4. Do i use 105 degree C, solid or stranded wire, 18 or 20gauge 600volt.
6.I want to use eyelet circuit board system, The standard thickness is 1/16" Where can i buy the eyelets and flaring tool. I would like to possibly use 1/8" material is there eyelets available.
I. know where to buy the board material
Q1: Yes - all tubes pin numbers, count clockwise when looking at the pins (i.e. from the bottom) starting at the reference/alignment tab.
Q2: Use Polypropylene Film/Foil where they will physically fit. For larger values then you just about have to use electrolytics unless you have a huge chassis to accommodate motor run style capacitors. Always use 105 degree C rated electrolytics where possible (a tube amp is a hot environment). Bypass electrolytics with (connect in parallel with) a series combination of a polyester or better yet a polypropylene capacitor of around 1/100th the value of the electrolytic and a 2.2 to 4.7 Ohm resistor.
(Explaination: Every cap has a self resonant frequency. Above that frequency they act like inductors not capacitors. Electrolytics can have quite low self resonant frequency, in the audio band even. An inductor and a capacitor in parallel can ring badly - because they form a tuned circuit. The resistor in series with the parallel polypropylene capacitor will damp any ringing which might occur above the electrolytics self resonant frequency - that is when it acts like an inductor)
Q3: Once again every man has his preferences. I use 2W Metal Film everywhere except for:
Grid stop resistors - use carbon film
High Power (where 2 W is not enough) use Wellwyn or similar Wirewound.
Q4: addressed above. Also when using electrolytic capacitors try to use them at around 70 to 75% of their rated voltage. That is, for example, using a 450V rated capacitor on a 430V rail is BAD design. For the example 430V rail I would use a 525V rated electrolytic or more likely series connect two 285V rated capacitors with voltage sharing resistors across them.
Q5: There was no Q5
Q6: London Power has eyelet boards, flaring tools etc. You can buy a starter kit of board,eyelets and a tool or top up packs, check their web pages. - Would you believe that here in Australia I order my eyelet board materials from Kevin there in Canada.
Hope this is of help.
Cheers,
Ian
Q2: Use Polypropylene Film/Foil where they will physically fit. For larger values then you just about have to use electrolytics unless you have a huge chassis to accommodate motor run style capacitors. Always use 105 degree C rated electrolytics where possible (a tube amp is a hot environment). Bypass electrolytics with (connect in parallel with) a series combination of a polyester or better yet a polypropylene capacitor of around 1/100th the value of the electrolytic and a 2.2 to 4.7 Ohm resistor.
(Explaination: Every cap has a self resonant frequency. Above that frequency they act like inductors not capacitors. Electrolytics can have quite low self resonant frequency, in the audio band even. An inductor and a capacitor in parallel can ring badly - because they form a tuned circuit. The resistor in series with the parallel polypropylene capacitor will damp any ringing which might occur above the electrolytics self resonant frequency - that is when it acts like an inductor)
Q3: Once again every man has his preferences. I use 2W Metal Film everywhere except for:
Grid stop resistors - use carbon film
High Power (where 2 W is not enough) use Wellwyn or similar Wirewound.
Q4: addressed above. Also when using electrolytic capacitors try to use them at around 70 to 75% of their rated voltage. That is, for example, using a 450V rated capacitor on a 430V rail is BAD design. For the example 430V rail I would use a 525V rated electrolytic or more likely series connect two 285V rated capacitors with voltage sharing resistors across them.
Q5: There was no Q5
Q6: London Power has eyelet boards, flaring tools etc. You can buy a starter kit of board,eyelets and a tool or top up packs, check their web pages. - Would you believe that here in Australia I order my eyelet board materials from Kevin there in Canada.
Hope this is of help.
Cheers,
Ian
Kt88 and HAmmond Circuits
Thanks Ian for your help
I recieved my POP book today and checked out the circuits you mentioned. Thank again
More questions
1 The curcuits are designed with 6550 for the 50, 60, 100 &120 watt amps. I would like to use KT 88 tube wich have higher values.Is this possible and what are the downfalls.
2. The circuits are for mono block design. I need to build stereo chassis and need clarifications. Is the input transformer designed and have enough ma to drive both channels? Secondly The same question aplies to the filament winding is there enough current?
3. Has anyone build these circuits.
4. What is the wire selection stranded or solid 18 or 20 gauge.
5. I keep forgetting 5
6. Can some of the diodes be replaced with bridget rectifiers.
7. There is no gain stage in the design. I would like to add this how and where. If you can help please put it into Idiots terms
Thanks
Thanks Ian for your help
I recieved my POP book today and checked out the circuits you mentioned. Thank again
More questions
1 The curcuits are designed with 6550 for the 50, 60, 100 &120 watt amps. I would like to use KT 88 tube wich have higher values.Is this possible and what are the downfalls.
2. The circuits are for mono block design. I need to build stereo chassis and need clarifications. Is the input transformer designed and have enough ma to drive both channels? Secondly The same question aplies to the filament winding is there enough current?
3. Has anyone build these circuits.
4. What is the wire selection stranded or solid 18 or 20 gauge.
5. I keep forgetting 5
6. Can some of the diodes be replaced with bridget rectifiers.
7. There is no gain stage in the design. I would like to add this how and where. If you can help please put it into Idiots terms
Thanks
Hi jbldriven,
You could take a shortcut and look up the typical application data for 6550 and for KT88 here
http://www.duncanamps.com/
You will see that the application data is identical. That is the rail voltages, tube currents, transformer primary Raa impedances etc. are all the same. The difference is that the KT88 can handle more current, dissipate more power and handle higher voltages and so should last longer than a 6550 in the same circuit.
Now that you have your copy of POP then you have lots of usefull info. You can look at the Transformer driven Design Procedure on page 5-9. All the data you need to apply that procedure is in the table at the top of page 5-13 and in the Appendix C (page C-2) at the back of the book.
As a worked examples for the Hammond 1650N and 1650P
1/ The 1650N
Raa = 4300 Ohms
Ra = Raa/4 = 1075 Ohms
Peak Drive Volts = root(2 x Power Out x Ra) = 359.2 V pk
Peak Drive Current = Vpk/Ra = 334.1mA
For Class AB1:
Va = Vpk + Vsat = 359.2 + 90 = 450V (Vsat = 90V for 6550 or KT88)
Pa = (450 x 450) / 4300 = 47 Watts total
Looking at the table on page 5-13
2 x EL34 marginal on power rating and grid2 voltage rating
2 x 6L6GC would be OK
Note: EL34 and 6L6 Vsat is only 50V - need 40V less High Volts.
2 x 6550 marginal on grid2 voltage rating
2 x KT88 would be being run really conservatively
High Voltage Rail of 450 V min @ 250mA for one channel
Repeat the above for the 1650P
Raa = 6600 Ohms
Ra= 1650 Ohms
Vpk = 445
Ipk = 269.7 mA
Va = 535V (for 6550 or KT88)
Pa = 43.4 Watts total
In this case ONLY the 6L6GC or KT88 would be advisable because of the high voltage on the ultralinear taps
High Voltage Rail of 550V min @ 200mA for one channel
Q1: Folk abuse tubes all the time particularly with respect to their voltage ratings BUT in this case not only is it possible to simply substitute KT88 for the 6550 it is actually advisable because the higher voltage ratings of the KT88 are (strictly speaking) required.
There are no downfalls and there is in fact one additional benefit the above doesn't show. The KT88 has a gm of 12 vs 10 for the 6550. That will actually make them a little easier to drive, so the driver stage doesn't have to work quite as hard.
Q2: Yes the circuits given are for mono blocks. If you build them as a stereo chassis then you will have to get onto the Hammond website and see if you can get a power transformer with double the High Voltage and Heater winding currents or jus build 2 monoblocks on a single chassis complete with separate power supplies. I would advise building them as monoblocks - at least you'll be able to lift them. As a stereo chassis remember your lifting drills from work - bend your knees etc. or you'll put your back out.
Each Monoblock:
See above for HV needs
Heaters 4A for each channel is OK
Q3: Kevin O'Connor undoubtably has. I haven't built either of those but built a version of the 120W with Hammond 1650T on page 5-33. I found the voltage was too high for EL34s (got flash overs in the tubes) and ended up with guess what?? - KT88s
Q4: The popular wisdom (this week at least) seems to be solid core and thin (ie 20 AWG in preference to 18AWG) for high voltage adn signal wires. See a post elswhere here regards earthing / 0V wiring. I would'nt be too fussed about heater wiring. Stranded or solid core will do - wire each channel separately - absolute minimum of 22 AWG for heater wires (22 AWG is 5 amp rated).
Q5: I forgot
Q6:Not easily - the cetre Tapped Tranny really needs separate diodes.
Q7: Just add a common cathode gain stage on the front end. Can follow this up latter if required - another 12A?7 will take heater current up to 4.1 Amps per channel.
Cheers,
Ian
You could take a shortcut and look up the typical application data for 6550 and for KT88 here
http://www.duncanamps.com/
You will see that the application data is identical. That is the rail voltages, tube currents, transformer primary Raa impedances etc. are all the same. The difference is that the KT88 can handle more current, dissipate more power and handle higher voltages and so should last longer than a 6550 in the same circuit.
Now that you have your copy of POP then you have lots of usefull info. You can look at the Transformer driven Design Procedure on page 5-9. All the data you need to apply that procedure is in the table at the top of page 5-13 and in the Appendix C (page C-2) at the back of the book.
As a worked examples for the Hammond 1650N and 1650P
1/ The 1650N
Raa = 4300 Ohms
Ra = Raa/4 = 1075 Ohms
Peak Drive Volts = root(2 x Power Out x Ra) = 359.2 V pk
Peak Drive Current = Vpk/Ra = 334.1mA
For Class AB1:
Va = Vpk + Vsat = 359.2 + 90 = 450V (Vsat = 90V for 6550 or KT88)
Pa = (450 x 450) / 4300 = 47 Watts total
Looking at the table on page 5-13
2 x EL34 marginal on power rating and grid2 voltage rating
2 x 6L6GC would be OK
Note: EL34 and 6L6 Vsat is only 50V - need 40V less High Volts.
2 x 6550 marginal on grid2 voltage rating
2 x KT88 would be being run really conservatively
High Voltage Rail of 450 V min @ 250mA for one channel
Repeat the above for the 1650P
Raa = 6600 Ohms
Ra= 1650 Ohms
Vpk = 445
Ipk = 269.7 mA
Va = 535V (for 6550 or KT88)
Pa = 43.4 Watts total
In this case ONLY the 6L6GC or KT88 would be advisable because of the high voltage on the ultralinear taps
High Voltage Rail of 550V min @ 200mA for one channel
Q1: Folk abuse tubes all the time particularly with respect to their voltage ratings BUT in this case not only is it possible to simply substitute KT88 for the 6550 it is actually advisable because the higher voltage ratings of the KT88 are (strictly speaking) required.
There are no downfalls and there is in fact one additional benefit the above doesn't show. The KT88 has a gm of 12 vs 10 for the 6550. That will actually make them a little easier to drive, so the driver stage doesn't have to work quite as hard.
Q2: Yes the circuits given are for mono blocks. If you build them as a stereo chassis then you will have to get onto the Hammond website and see if you can get a power transformer with double the High Voltage and Heater winding currents or jus build 2 monoblocks on a single chassis complete with separate power supplies. I would advise building them as monoblocks - at least you'll be able to lift them. As a stereo chassis remember your lifting drills from work - bend your knees etc. or you'll put your back out.
Each Monoblock:
See above for HV needs
Heaters 4A for each channel is OK
Q3: Kevin O'Connor undoubtably has. I haven't built either of those but built a version of the 120W with Hammond 1650T on page 5-33. I found the voltage was too high for EL34s (got flash overs in the tubes) and ended up with guess what?? - KT88s
Q4: The popular wisdom (this week at least) seems to be solid core and thin (ie 20 AWG in preference to 18AWG) for high voltage adn signal wires. See a post elswhere here regards earthing / 0V wiring. I would'nt be too fussed about heater wiring. Stranded or solid core will do - wire each channel separately - absolute minimum of 22 AWG for heater wires (22 AWG is 5 amp rated).
Q5: I forgot
Q6:Not easily - the cetre Tapped Tranny really needs separate diodes.
Q7: Just add a common cathode gain stage on the front end. Can follow this up latter if required - another 12A?7 will take heater current up to 4.1 Amps per channel.
Cheers,
Ian
KT88 Hammon 60 watt 1650p 278c
Thanks Ian for your support. I have information sent from London power answering the same similar questions. They stated i could use the 278cx in place to provide higher ma to support stereo. He didin't specify any change to the caps down stream in the circuit to support the second channel. Rather he recommended the following:
Question;
I need to build a stereo amp on one chassis. What are the power supply changes that are required to support the current requirements for 2 channel instead of mono block. Does Hammond have the transformer required --278cx?. Would it be feasible to use the power rectification stage from 120 watt circuit? If so what would the circuit look like.
Answer;
Just sub the 278CX PT.
Or
Personally, I would use the splitter from page 6-14 and eliminate the negative rail for the splitter. You can retain the direct-connected cathode followers if you like.
Can you help provide me with your thoughts on this
Thanks Ian for your support. I have information sent from London power answering the same similar questions. They stated i could use the 278cx in place to provide higher ma to support stereo. He didin't specify any change to the caps down stream in the circuit to support the second channel. Rather he recommended the following:
Question;
I need to build a stereo amp on one chassis. What are the power supply changes that are required to support the current requirements for 2 channel instead of mono block. Does Hammond have the transformer required --278cx?. Would it be feasible to use the power rectification stage from 120 watt circuit? If so what would the circuit look like.
Answer;
Just sub the 278CX PT.
Or
Personally, I would use the splitter from page 6-14 and eliminate the negative rail for the splitter. You can retain the direct-connected cathode followers if you like.
Can you help provide me with your thoughts on this
KT88 hammond circuit 60 watt 1650p
Ian can you also review the question for the bias curcuit. Here are there recommended changes. Do you have the capabilities to email a sketch overlaying the change ( if it is good )
Question;
There are 2 variable pots on the KT88 bias circuit. One is 22kOhm and the other is ??? What are the specifics on these pots such as voltage watts.
Answer
1/2W pots are fine, the 16mm types work okay.
or
A better bias circuit is to change the 160k resistors to 100k-2W flameproofs, then feed to the first cap as shown, but reduce the 47uFs to 10uF each. Use a pair of 50k linear pots wired in parallel with each wiper feeding one of the 100k grid-leak resistors for the output tubes. The ground-end of the pots then ties to ground through a 6k8 to 10k 1/2W resistor.
Secondly you said that you build the 120 watt version. This is also my intention and i would like to puchase the part at the same time. Did you make any mods. I will make them mono blocks
Could you send me some pictures and by chance would you have a complete parts list taken off from the drawing.
Ian can you also review the question for the bias curcuit. Here are there recommended changes. Do you have the capabilities to email a sketch overlaying the change ( if it is good )
Question;
There are 2 variable pots on the KT88 bias circuit. One is 22kOhm and the other is ??? What are the specifics on these pots such as voltage watts.
Answer
1/2W pots are fine, the 16mm types work okay.
or
A better bias circuit is to change the 160k resistors to 100k-2W flameproofs, then feed to the first cap as shown, but reduce the 47uFs to 10uF each. Use a pair of 50k linear pots wired in parallel with each wiper feeding one of the 100k grid-leak resistors for the output tubes. The ground-end of the pots then ties to ground through a 6k8 to 10k 1/2W resistor.
Secondly you said that you build the 120 watt version. This is also my intention and i would like to puchase the part at the same time. Did you make any mods. I will make them mono blocks
Could you send me some pictures and by chance would you have a complete parts list taken off from the drawing.
jbldriven,
Sorry about the tardy reply. I am away from home for work for the next 4 weeks. While I still have some email access (at crappy dial up data rate) I don't have access to all my reference books and notes.
Kevins advice as noted above is good. Is the phase splitter he suggests (the one from page 6-14) the 12AX7 diffamp with transistor current source in the cathods driven via a pair of 1M resistors off the 2 triode anodes. If so then that is a good suggestion. That design uses AC feedback from the triode outputs (anodes) to the current source to ensure good AC balance. You will need the direct coupled cathode followers (as shown on one of the designs you reference above) to drive the KT88 grids. The bias circuit described is also a good idea.
Sorry I can't be more specific but I just don't have my copies of the London Power Books with me.
My 120W amp was a bit different. It uses a CAE PCB fro the front end which is exactly like the suggested diffamp/phase splitter above but uses cascode connected 6DJ8 on each side of the diffamp instead of a 12AX7 single triode half on each side. Then uses direct coupled 6SN7 cathode followers to drive two pairs of KT88 in Ultralinear Mode with a Hammond 1650T - So its a "dressed up" version of what Kevin suggests. Put that aside until you get this project done thou'. If you want a bit of a look, or to save schametic for future reference then look here:
www.evatco.com.au
Click on the projects page tab on the left hand side.
The prototype is in the 3rd example project down. There are amp and power supply schematics and a bit of a rave document about what I was trying to do. The amp schemaic shows a high voltage regulator which contains a zener diode. This zener diode is drawn backward on the schematic else the circuit is correct. The design has been discussed here on a couple of occassions with a few folk saying that they would do it differently if it were them but no one saying thats rubbish and won't work. Actually I have a prototype running so I know it does work and sounds really good (in mono at least).
Cheers,
Ian
Sorry about the tardy reply. I am away from home for work for the next 4 weeks. While I still have some email access (at crappy dial up data rate) I don't have access to all my reference books and notes.
Kevins advice as noted above is good. Is the phase splitter he suggests (the one from page 6-14) the 12AX7 diffamp with transistor current source in the cathods driven via a pair of 1M resistors off the 2 triode anodes. If so then that is a good suggestion. That design uses AC feedback from the triode outputs (anodes) to the current source to ensure good AC balance. You will need the direct coupled cathode followers (as shown on one of the designs you reference above) to drive the KT88 grids. The bias circuit described is also a good idea.
Sorry I can't be more specific but I just don't have my copies of the London Power Books with me.
My 120W amp was a bit different. It uses a CAE PCB fro the front end which is exactly like the suggested diffamp/phase splitter above but uses cascode connected 6DJ8 on each side of the diffamp instead of a 12AX7 single triode half on each side. Then uses direct coupled 6SN7 cathode followers to drive two pairs of KT88 in Ultralinear Mode with a Hammond 1650T - So its a "dressed up" version of what Kevin suggests. Put that aside until you get this project done thou'. If you want a bit of a look, or to save schametic for future reference then look here:
www.evatco.com.au
Click on the projects page tab on the left hand side.
The prototype is in the 3rd example project down. There are amp and power supply schematics and a bit of a rave document about what I was trying to do. The amp schemaic shows a high voltage regulator which contains a zener diode. This zener diode is drawn backward on the schematic else the circuit is correct. The design has been discussed here on a couple of occassions with a few folk saying that they would do it differently if it were them but no one saying thats rubbish and won't work. Actually I have a prototype running so I know it does work and sounds really good (in mono at least).
Cheers,
Ian
kt88 HAmmond Stat up
Gingertube
I have assembled the 60 watt per channel kt 88 amp and presently experiencing a few problem. I was wondering if they are similar to your start up.
I have a slight buzz out of one channel and the other is substaintially louder. The buzz does not get louder as the volume increases.
I ran the unit oon a scope with 1000 hz. I could not input more than 2 volts before the second half of the drive tube started to distort on half of the wave. I meassureed the output and was only getting about 5 watts pior to distortion.
I use the transistor balance cicuit as desribed in the London Power book. It was suggested over the negative rail. I am using hammold 2- 1650p and 1- cx278. I have about 600 volts B+ and 540v V+. My Bias is ajuatable from -25 to -90 vdc.
What is the input votage normally for any amplifer.
I am using one common start ground. Shoud i try moving the transformers or isolatinfg the ground wires one by one. I was told the the transofmer field may be the problem. The high voltage tranny is mount at 90 degrees to both output with 25mm between.
I don't have enough overall gain and have to pumo it up by the electronic crossover +10 db.
I know you would have to look at the circuit to see and irregularities in values. If you need the ciruit i would like to email direct instead of posting.
Gingertube
I have assembled the 60 watt per channel kt 88 amp and presently experiencing a few problem. I was wondering if they are similar to your start up.
I have a slight buzz out of one channel and the other is substaintially louder. The buzz does not get louder as the volume increases.
I ran the unit oon a scope with 1000 hz. I could not input more than 2 volts before the second half of the drive tube started to distort on half of the wave. I meassureed the output and was only getting about 5 watts pior to distortion.
I use the transistor balance cicuit as desribed in the London Power book. It was suggested over the negative rail. I am using hammold 2- 1650p and 1- cx278. I have about 600 volts B+ and 540v V+. My Bias is ajuatable from -25 to -90 vdc.
What is the input votage normally for any amplifer.
I am using one common start ground. Shoud i try moving the transformers or isolatinfg the ground wires one by one. I was told the the transofmer field may be the problem. The high voltage tranny is mount at 90 degrees to both output with 25mm between.
I don't have enough overall gain and have to pumo it up by the electronic crossover +10 db.
I know you would have to look at the circuit to see and irregularities in values. If you need the ciruit i would like to email direct instead of posting.
It may be that Hammond 2XX power tranny. Everytime I've used one of the 2XX series trannies I have had horrible hum problems. The only way I could eliminate the hum was to build seperate PS chassis. If you plan on making more amps I would suggest you source a power tranny that is higher quality or get some custom wound toroidal's.
good luck.
good luck.
JBLdriven
You said
"I ran the unit oon a scope with 1000 hz. I could not input more than 2 volts before the second half of the drive tube started to distort on half of the wave. I meassureed the output and was only getting about 5 watts pior to distortion. "
This is the transistor current source in the commoned cathodes of the input diffamp running out of compliance voltage. In other words, the input stage cannot handle the signal swing required to drive the amp to full power. You either have to add more gain after the input stage to reduce the required input signal swing or return the bottom of the current source to a negative rail (the bias supply for example) to allow it to handle more signal swing.
On the hum issue - does the heater winding have a centre tap and if so have you connected it to earth. If not then connect a 47R 1W resistor from each side of the heater winding to earth.
Hope this helps.
Cheers,
Ian
You said
"I ran the unit oon a scope with 1000 hz. I could not input more than 2 volts before the second half of the drive tube started to distort on half of the wave. I meassureed the output and was only getting about 5 watts pior to distortion. "
This is the transistor current source in the commoned cathodes of the input diffamp running out of compliance voltage. In other words, the input stage cannot handle the signal swing required to drive the amp to full power. You either have to add more gain after the input stage to reduce the required input signal swing or return the bottom of the current source to a negative rail (the bias supply for example) to allow it to handle more signal swing.
On the hum issue - does the heater winding have a centre tap and if so have you connected it to earth. If not then connect a 47R 1W resistor from each side of the heater winding to earth.
Hope this helps.
Cheers,
Ian
KT 88 Start Up Corrections
THanks Gingertube
I am using the 278cx tranny heater circuit to heat the 4 kt 88 tubes (parrallel) with no resistors. The drive tube 2-12AX7 and the 2- 12At7 (parrallel) are driven with a seperate 12 volt tranny complete with 100r 1w resistors in serries on each leg to ground.
I have been talking to London Power to find a solution. I think Kevin is away. Could it be a problem with the feedback, possibly a resistor value change. The transistor ciruit is used to provide better balance as Kevin had mentioned.
The original design from Page 5-30 shows negative rail supply however I was advised to use the 6-3 splitter on page 6-14.
I am using a 33k ressistor in place of the 3k on the feedback circuit to the pin 7 of the drive tube. The 100r ressistor to ground is replaced with a 1 k resistor. I am waiting to talk to kevin regaurding the problem. I'm sure there will be a simple cure.
Thoughts?
As for the buzz it is primarly in the one channel. All heater wiring is twisted. Someone mentioned that i may have to move the output slightly to counter the fields. This may be as little 5mm.
Have you heard of this before
THanks Gingertube
I am using the 278cx tranny heater circuit to heat the 4 kt 88 tubes (parrallel) with no resistors. The drive tube 2-12AX7 and the 2- 12At7 (parrallel) are driven with a seperate 12 volt tranny complete with 100r 1w resistors in serries on each leg to ground.
I have been talking to London Power to find a solution. I think Kevin is away. Could it be a problem with the feedback, possibly a resistor value change. The transistor ciruit is used to provide better balance as Kevin had mentioned.
The original design from Page 5-30 shows negative rail supply however I was advised to use the 6-3 splitter on page 6-14.
I am using a 33k ressistor in place of the 3k on the feedback circuit to the pin 7 of the drive tube. The 100r ressistor to ground is replaced with a 1 k resistor. I am waiting to talk to kevin regaurding the problem. I'm sure there will be a simple cure.
Thoughts?
As for the buzz it is primarly in the one channel. All heater wiring is twisted. Someone mentioned that i may have to move the output slightly to counter the fields. This may be as little 5mm.
Have you heard of this before
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.