Princeton Reverb Preamp, Phase Inverter to ICEPower amp

[sluckey]

Mojo779 or Weber WRVBPT are small and will make 330 to 350VDC with a FWB. Simple and straight forward. You can buy two of the Weber PTs for less than the Mojo779. I have the Weber in my revibe...

     https://sluckeyamps.com/revibe/revibe.pdf

[dbishopbliss]

I have the Weber in my revibe...
     https://sluckeyamps.com/revibe/revibe.pdf

Did you notice that Weber references your ReVibe on their site? Maybe it's not yours. I notice you credit Hoffman.

[tubeswell]

There's revibes and there's revibes. Weber VST credits their kit to a Jeff Gehring design. https://www.tedweber.com/5h15-c-kt/

[sluckey]

Did you notice that Weber references your ReVibe on their site? Maybe it's not yours. I notice you credit Hoffman.

I've never seen my name on Weber's site. Have you? I'm unsure of the origins of the revibe. I first saw the revibe on Hoffman's site. I used his schematic to build mine but designed my own layout. I also made some modifications, mostly to deal with hot active pickups.

[dbishopbliss]

I've never seen my name on Weber's site. Have you? I'm unsure of the origins of the revibe. I first saw the revibe on Hoffman's site. I used his schematic to build mine but designed my own layout. I also made some modifications, mostly to deal with hot active pickups.

I always associated it with you because I saw it on your site.

[dbishopbliss]

Mojo779 or Weber WRVBPT are small and will make 330 to 350VDC with a FWB. Simple and straight forward. You can buy two of the Weber PTs for less than the Mojo779.

I updated my power supply using a FWB based on the Weber WRVBPT. I picked 40uF for the filter caps because it made the curves smoother in the simulation and CE MFG makes a 40/40/40 multi-section cap. Although, I might go with separate caps depending upon the enclosure size and cost savings. What are the advantages/disadvantages with going larger or smaller. I know the Bandmaster uses 50uF.


image upload

[tdvt]

Sorry, another side query.

Looking at your PSU image, I see you are using the constant current selection as the load. Are you getting pretty good ballpark results?

I am typically unsure about where to go with that choice, & I seem to recall PRR once suggesting a 5K resistive load was an OK general starting point.

[sluckey]

Since you are building a preamp only I suggest big filter caps for better smoothing. Smoother is better IMO. I'd probably use two 100µF for the first two nodes.

[dbishopbliss]

Looking at your PSU image, I see you are using the constant current selection as the load. Are you getting pretty good ballpark results?

I am typically unsure about where to go with that choice, & I seem to recall PRR once suggesting a 5K resistive load was an OK general starting point.

I'm not sure... I thought using the current of the tubes would be a good way to estimate. I will trying swapping in a 5K resistive load and see what happens. I wonder where the 5K value comes from.

[tdvt]

I would have thought current would be the first choice myself, but I seem to recall discussion of things being more dynamic/complicated than was worth going that way.


Don't quote me on the 5K, but that is in the back of my mind.

I was hoping you had it down already...


I wish I could remember the thread, I will have to hunt around when I have more time.

[sluckey]

Are you just having fun using the power supply software? Or would you just like to quickly get some real world numbers?

The preamp and PI tubes will only draw about 4mA from a 224VDC source. Look at my 5E3 schematic...

     https://sluckeyamps.com/5e3/5e3.pdf

You can calculate the current requirement by dividing the individual cathode voltages by the cathode resistance then adding all those currents together. Or, an easier way is simply subtract node C voltage from node B voltage and divide by 22K. Either way gives about 4mA.

[tdvt]

When I use that program, I am mostly looking to ballpark my 1st node voltage (generally close to correct), but also my 4th node voltage (usually not so correct).

That is why I was asking how close his numbers have been.

[dbishopbliss]

Are you just having fun using the power supply software? Or would you just like to quickly get some real world numbers?

I changed the subject because I am using a Princeton Reverb preamp instead of the 5E3 because I only want one channel and I want the extra gain stage of the reverb circuit without the actual reverb. The other change is that I am using a Fixed Bias Cathodyne Phase Inverter using a 12AU7 (at least that half of a 12DW7). So I can use the real world numbers from the Fender schematic for the first 3 gain stages but the phase inverter is different. That is why I'm trying to calculate what is the expected current draw and use that to determine the voltage dropping resistor values for the power supply.

The preamp and PI tubes will only draw about 4mA from a 224VDC source. Look at my 5E3 schematic...     https://sluckeyamps.com/5e3/5e3.pdfYou can calculate the current requirement by dividing the individual cathode voltages by the cathode resistance then adding all those currents together. Or, an easier way is simply subtract node C voltage from node B voltage and divide by 22K. Either way gives about 4mA.

I'm using the following for my calculations based on the schematic below: 1.3V / 1500R = 0.86mA * 3 = 2.6mA. My modified Phase Inverter will have 107V / 5.6K = 19mA.Do I have that correct or am I missing something?

[dbishopbliss]

When I use that program, I am mostly looking to ballpark my 1st node voltage (generally close to correct), but also my 4th node voltage (usually not so correct).

That is why I was asking how close his numbers have been.

I will have to let you know when I build it.

[tubeswell]

I think it could work, but after pondering the PT load, even though your little PT 125-0-125 winding has a design max of 25mA, getting 20 or so mA out of it might make it sag a bit, and, running quite warm it will have a shorter life expectancy. And in a pedal case, it will dissipate quite a bit of heat, so will the AU triode. So you may want to get a bigger pt (and a bigger pedal case) and have some forced ventilation cooling in your pedal. BTW, this concept is quite experimental so I would breadboard it first (as I think I suggested at the beginning). But don’t give up. I love mad science.


Edit: Also remember that those 5k6 load resistors will want to be at least 5W each (preferably bigger if everything is crammed into a pedal case).

[tubeswell]

Also, there are a couple of alternative ways to get the cathodyne bias.

One is designing the preceding triode to have a plate idle voltage of around 100v and do DC coupling to the cathodyne grid. This would eliminate the need for the fixed bias resistors and coupling cap, but would mean you need to adjust the plate supply and plate load etc of the driving stage to get the plate idling at between 100 and 110V. (But breadboard it with the fixed bias resistors first to ensure your cathodyne triode behaves as expected. If you go to DC coupling, the on-paper voltages will shift around a bit due to current ‘stealing’ especially because the cathodyne will draw about 20x the current of the driving stage if you keep the 12DW7 - so this idea may need a bit of additional experimentation, and may result in a weird output signal with a 12DW7. But it could work if you use a 12AU7 for the driving stage as well, if the driver is set up with a 10k plate load maybe, which would further increase the current draw on the HT winding).

The other alternative is that with the fixed bias, the input impedance using the voltage divider will be 2M2||1M = 687.5k , which is okay but not super high. By adding another 1M resistor between the voltage divider output and the junction of the coupling cap and the cathodyne grid, together with a decoupling cap from the output of the voltage divider to ground, you can increase the input impedance to 1M. This is better but don’t know if necessary until you get it working. (But it will work with the 12DW7 driver)

[dbishopbliss]

I think it could work, but after pondering the PT load, even though your little PT 125-0-125 winding has a design max of 25mA, getting 20 or so mA out of it might make it sag a bit, and, running quite warm it will have a shorter life expectancy. And in a pedal case, it will dissipate quite a bit of heat, so will the AU triode. So you may want to get a bigger pt (and a bigger pedal case) and have some forced ventilation cooling in your pedal. BTW, this concept is quite experimental so I would breadboard it first (as I think I suggested at the beginning). But don’t give up. I love mad science.


Edit: Also remember that those 5k6 load resistors will want to be at least 5W each (preferably bigger if everything is crammed into a pedal case).

I've been trying to come up with a layout for a 12" x 8" enclosure, but I'm not sure I can make it all fit with the tubes and transformer inside so I will either have to go with a larger chassis or go for a more traditional layout with transformer and tubes coming out the top. Maybe this is why you don't see tubes in pedal board amps. Well at least 2 tubes. Milkman The Amp has a single 12AX7 - which is interesting because it does not have any ventilation in the enclosure. If I recall they are using the same ICEPower module that I am using.

[tubeswell]

Milkman The Amp has a single 12AX7 - which is interesting because it does not have any ventilation in the enclosure. If I recall they are using the same ICEPower module that I am using.

Why not just copy what they do? (Or just buy one?)

[tdvt]

I built a Champ+2-tube reverb in a 7 X 11 chassis, but it was not all inside as you are trying to do.  8 x 12 isn't really big, but hard to guess at not having actually seen the module.

Have you considered maybe a mesh tube-cage (B-15 style) on top of part of the chassis for space & ventilation? The tubes could even only partially protrude to keep a lower profile, but it would certainly help with any heat.

I have also used a few Antek toriodal PTs. They are a somewhat larger footprint (diameter) than a traditional PT but are fairly short, which might offer more design options. They will fit (barely) on the inside of a standard Fender chassis which is 1-3/4" deep.

Also, looking back through this thread but didn't see it mentioned, I was wondering about a matching transformer for the line out?   

[dbishopbliss]

Milkman The Amp has a single 12AX7 - which is interesting because it does not have any ventilation in the enclosure. If I recall they are using the same ICEPower module that I am using.

Why not just copy what they do? (Or just buy one?)

What's the fun in that? Plus I already have the board. :-D

[tdvt]

Your project has been food for thought in mixing SS & tube components.

I was flashing back on a few shows I worked where the bands were carrying Fender Tone Master amps (the newer modelling type) &/or Quilters & remembering that they didn't sound bad, especially in the context of a band. They liked them because the were loud & light.

My bass amps have been hybrids for years & years.

The SS power block concept might be especially applicable to a stereo combo, something that comes up regularly at the moment.

Not much technical info on the Fender TM's but this pic was interesting.

[dbishopbliss]

Not much technical info on the Fender TM's but this pic was interesting.

That looks like the 100AS1 board.

[dbishopbliss]

I ordered the power transformer from Weber and the other parts I needed over the weekend. Next up I will work on the turret board layout. I have a feeling this amp will end being a "lunch box" amp instead of a pedal board amp. Although the first build will probably be in a single flat piece of aluminum I have.


One other thought that occurred to me after I ordered the parts. Instead of using a Tube Phase Inverter to split the signal, I think I could I have used an interstage transformer like Bartel amps. Although, I'm not sure that would have really saved me any space in the Princeton Reverb circuit because of the 3 gain stages. I would have to simplify to use a single tube and could have used the transformer I already own.


Any thoughts on how to use a transformer to create the balanced signal? I can always breadboard both and just swap the amp board between them.

[tdvt]

Any thoughts on how to use a transformer to create the balanced signal? I can always breadboard both and just swap the amp board between them.

I don't know enough about that stuff to advise but I would have thought something along the lines of the attached diagram.

I apparently skimmed right over your mention of the line transformer early on.

I bought a couple Edcor line transformers for a pre-amp project but they are still in the drawer, so no experience to share.

[dbishopbliss]

I don't know enough about that stuff to advise but I would have thought something along the lines of the attached diagram.

I apparently skimmed right over your mention of the line transformer early on.

I saw that early on, but the cost ($100+) was a deterrent so I didn't pursue it. Perhaps the cost wasn't as much as I thought compared to buying a new power transformer, additional capacitors and resistors and a 12DW7 tube. Especially when you consider the shipping costs added on top.


But looking at Edcor, maybe the WSM10K/10K would work. I think this is the same thing but with solder tabs instead of wires.

[dbishopbliss]

I'm ordering parts and it is harder than expected to find 5W resistors in the values I am looking for. Well... maybe not harder, but they don't have them stocked at Mouser which is where I usually order parts from. I decided to check Newark and saw NEOHM - TE CONNECTIVITY ROX5SSJ5K6. Which is a "Super Small" version of their 5W Resistor. While not a HUGE difference, smaller is usually easier when laying out the parts. But what really distinguished this part is the price... Less than thirteen cents ($0.125). The Vishay brand of the same value and power rating is $2.61.


Any reason to NOT use the cheaper version?

[acheld]

Any reason to NOT use the cheaper version?

No reason at all.  Do check the resistance.   I've seen some variance from TE Connectivity (maybe 1 in 15 off spec). 

[tubeswell]

I’ve found 5W ones on eBay often. If it’s difficult to get them ATM, why not look for 10W ones? Or just cobble a string of 1k 2W resistors together (for breadboarding).

[dbishopbliss]

I’ve found 5W ones on eBay often. If it’s difficult to get them ATM, why not look for 10W ones? Or just cobble a string of 1k 2W resistors together (for breadboarding).

I am assuming (which may be a bad thing) that the basic circuit topology is correct so I'm laying out a turret board and assuming I can swap resistor values as needed. Since I'm hoping for a compact board I am trying to avoid a bunch of large resistors. Not to mention... $0.12 per resistor is pretty cheap. At least cheap enough that I don't care about the cost. I just bought a taco that cost $3.00. :-D

[dbishopbliss]

Here's the board and the schematic. All components are to scale. Grid Stoppers are off the board. I just realized I omitted the 10K resistors after the coupling caps going to ground before the ICEPower board. I'm actually not sure where I got those values.


Comments welcome and appreciated.

[tubeswell]

Does the ICE power module have its own input resistors? (if so, there's no need for resistors after the output of the cathodyne, but if you were inclined to put some in, I'd make them in the order of 1M each)


Edit - looking back at earlier replies and glancing at the datasheet, I see there is an input resistor for each input, so …

[dbishopbliss]

Edit - looking back at earlier replies and glancing at the datasheet, I see there is an input resistor for each input, so …

Thanks. I will remove them from the schematic and back out my update to the board.

[dbishopbliss]

And so it begins... I realize I should have probably spaced out the components horizontally a little more. Things are going to be tight.

[dbishopbliss]

Board done... realized one thing. I should have drilled holes to mount the board on stand-offs before I started soldering. Oh well, I should be able to drill without too much trouble.


I have a 12"x12" piece of aluminum that I might try mounting the preamp, power amp, tube sockets, pots, jacks and transformer to. I have a pair of angle pieces that were made to install a tube socket horizontally. If I use those, then I could mount everything under the plate so that it could be used as a somewhat large pedalboard amp. I will have to see how that works.

[dbishopbliss]

Thoughts on layout?

[tubeswell]

brushless fan for the 12AU7?

[shooter]

What he said ^^^^^

My SS module was getting real hot inside the box, no fan, brought it outside n used the recommended heat-sink.

[dbishopbliss]

brushless fan for the 12AU7?

Are most fans for electronics brushless? I'm not seeing that as a spec on the listings at parts express. This one runs off 24VDC. Since the amp has a 25VDC Aux power it could be a good option. What do you think?



Update: Found my answer here. "DC Fans are always brushless... "

[dbishopbliss]

The problem with cleaning up your basement is that you don't remember where you put anything. I found the ICEPower board, but I realized the wiring harness is not with it. I can't find it anywhere so I ordered a new one, but they are on backorder. :-/


Could I connect the phase inverter outputs to a plug (one plug for each phase) and then use that as an input to a power amp (I have a LM3886 amp I made years ago)?


It's a stereo amp so the signals would be out of phase. But, if I hooked up one speaker reversing the poles then they be back in phase with each other, right?

[shooter]

Cap coupled, might get some loading. 


swapping speaker polarity should work


FWIW this is the module I use, have driven it with a couple different tube types, never a PI
hunt up your SS amps spec sheet, you don't want to have too hot a signal.

[dbishopbliss]

Cap coupled, might get some loading. 

I don't know what that means. Is it a good thing, bad thing, just a thing?

[shooter]

make sure there is a series cap between the PI n SS module, even then, you might get some impedance loading

[HotBluePlates]

I have the Weber in my revibe...
     https://sluckeyamps.com/revibe/revibe.pdf

Did you notice that Weber references your ReVibe on their site? Maybe it's not yours. I notice you credit Hoffman.

There's revibes and there's revibes. Weber VST credits their kit to a Jeff Gehring design. https://www.tedweber.com/5h15-c-kt/

I've never seen my name on Weber's site. Have you? I'm unsure of the origins of the revibe. I first saw the revibe on Hoffman's site. I used his schematic to build mine but designed my own layout. I also made some modifications, mostly to deal with hot active pickups.

I always associated it with you because I saw it on your site.

FWIW:

"ReVibe" is a product/kit that WeberVST sells that mixes a standalone reverb unit with a harmonic tremolo circuit.  Jeff Gehring (who I've interacted with on another forum) apparently helped Weber develop the circuit & implementation of the kit.  I'm unsure exactly when or why Hoffman offered a layout for the circuit, but Sluckey notes he modified that to suit his needs.

Kits & forums being what they are, original ideas flow around in ways that it's hard to trace back to the origin.