Off to a good start, Mark. I'm looking forward to seeing this come together. Take your time, but hurry up!
Mark wrote: ↑Sat Jun 22, 2019 5:37 am
Thanks for you advice Colossal, any thoughts about caps and resistors?
Well, I am not a purist at heart so not inclined to pursue NOS parts or anything. I have been using good quality carbon film resistors in the signal path and metal film resistors for at least the first stage's plate resistor.
I am starting my own stripped down AC30 variant (an even slower build).
Have you loaded up your MM Woden PT on the bench yet? I have been speaking with Patrick about some of the Haddon PTs and they do indeed sag quite a bit. I will have transformers coming late next week to test but I don't want to hijack.
I will go with traditional filter cap values apart from the first cap which will be 32uF from the rectifier, and 16uF on the choke. I've heard the amp sounds more "organic" with the traditional filter caps. The 32uF helps with hum issues the old AC-30's experienced.
Feel free to post your stuff on this thread, all ideas welcome.
Mark wrote: ↑Thu Jul 04, 2019 9:00 pm
I will go with traditional filter cap values apart from the first cap which will be 32uF from the rectifier, and 16uF on the choke. I've heard the amp sounds more "organic" with the traditional filter caps. The 32uF helps with hum issues the old AC-30's experienced.
In perusing AC schematics, the AC30CC2 shows a Smoothing switch which switches in an extra 22uF on both the Plate and Screen nodes. They note that as Modern at 44uF and 22uF for Vintage. Also, the CC2 is running higher voltages, similar to the Marshall 18/36W amps.
AC30CC2_PS.png
Feel free to post your stuff on this thread, all ideas welcome.
Excellent. Are you planning to run your power supply elements in series like the Fryer amps (32u-CHOKE-16u-10k-16u-22k-16u) or with the Phase Inverter and Preamp nodes branching off the Screen supply, in parallel, like the AC30/6, AC30/4, or AC15?
I am working up up a little spreadsheet to model some scenarios then will do some tests when the transformers get here. I'm basing my power supply on the PI and preamp branching off the screen supply. Also, planning to try both a 20H 460Ω choke and 10H 280Ω. The corner frequencies are quite low, even with the low capacitance.
You do not have the required permissions to view the files attached to this post.
I have a Trainwreck Rocket so this will have parallelled supplies from the screen node.
EDIT:What I find interesting is the fuse on the 5vac, the 6.3vac, and the centre tap of the secondary of the power transformer. The amp is also above chassis earth which is unusual, though if you are like Brian May or Status Quo, it's quite a handy feature.
Excuse my ignorance, but what are you expecting to hear from different choke values?
Mark wrote: ↑Fri Jul 05, 2019 3:00 am
Excuse my ignorance, but what are you expecting to hear from different choke values?[/b]
I want to see how the chokes affect the immediacy and feel of the amp under the fingers (if at all). I am designing around a 300V plate voltage, plus or minus. Pretty low this time around. Experimentally, I've been going back and forth between what'll call Marshall (345V plate, 315V screen) and (lower) Vox EL84 voltages (300/300) and am leaning toward the lower voltage sound for this next build. At a plate voltage of 295-300V, I'm estimating 19mA of screen current and 5V dropped across the 10H choke and 9V across the 20H. The characteristic impedance of the 20H choke is about 30% higher. PI and preamp nodes drop by an additional 4V with the 20H choke. There is slightly more squish to the grid curves at the lower screen voltage as well.
Colossal wrote: ↑Fri Jul 05, 2019 4:20 pmExperimentally, I've been going back and forth between what'll call Marshall (345V plate, 315V screen) and (lower) Vox EL84 voltages (300/300) and am leaning toward the lower voltage sound for this next build.
Man, I may have to eat this. Argh, choices! Both sound really, really good but different.
How are you pulling the screens down to 315VDC. From my experience the higher voltage will sound brighter but this doesn't appear to be an issue in an AC-30. The output valves are also within their specification at the lower voltage too which is a plus.
Mark wrote: ↑Sat Jul 06, 2019 12:21 am
How are you pulling the screens down to 315VDC. From my experience the higher voltage will sound brighter but this doesn't appear to be an issue in an AC-30. The output valves are also within their specification at the lower voltage too which is a plus.
The amp I'm testing with is 345V plates and 315V screens. The drop is 2k2 across a 32-32u cap can, standard 18W fair. Very punchy at those voltages. A little compressed (for my taste) perhaps but works wonderfully. The brightness appears more forward with a bit of a Pluh--- sound on the attack. Quite defined and separated. If I reduce the plates to about 300V and I think something like 290V on the screens it's a bit softer but more lush. Incredibly warm sounding, more relaxed and blended, if you will. The mids are a little wetter, which I like. I'm using a RS OT. Oh, I should mention the cab: A '69 slant 4x12 with two 8Ω Heritage G12Ms wired in series. The cab is detuned with one speaker removed from the upper left and one from the lower right (if looking from the front). Placing the speakers diagonally versus next to each other in the upper holes sounded better. Absolutely stunning sound with a Les Paul. Very piano like with very nice sustain. Quite wet sounding as well, which is what I'm after.
with clc filter: 32uf after ss rectifier, 32uf after choke and 33uf at screens. preamp 10uf except cathode follower which is 22 or 33uf
Hi Roe,
So your entire supply is seeing the choke. How do you find the feel and sound? I read an interview with Brian May's tech where he said that Brian discerned no difference in sound between SS rectified and tube rectified. But his amps are just capacitive input filtered going to the plates, not CLC, then to the plates and screens like your amp. I believe the Dave Petersen amps were CLC filtered.
What are your plate, screen, and cathode voltages? I read you like 50R for the bias resistor with your setup.
Petersen BMAC.png
You do not have the required permissions to view the files attached to this post.
I do have a 32uf connected directly to the rectificer and choke. the amp feels firm with a fast response, probably a little less sag than most ac30s. the choke has a little less resistance than a gz34 but ac30s draw so much current at idle with the 50R katode resistor that it does not matter too much really. the amp is extremely quiet with tight bass. I'd gladly compare it to other ac30s
Plate voltage is around 320 and screens a little lower with a 470R shared screen resistor in series with 100R individual resistors
Roe wrote: ↑Mon Jul 08, 2019 7:55 am
Plate voltage is around 320 and screens a little lower with a 470R shared screen resistor in series with 100R individual resistors
How was the bass response affected by the change in screen voltage?
Roe wrote: ↑Mon Jul 08, 2019 7:55 am
Plate voltage is around 320 and screens a little lower with a 470R shared screen resistor in series with 100R individual resistors
How was the bass response affected by the change in screen voltage?
Not much since I replace a choke with a 420R resistance with a 470R resistor. the CLC filter and increased filtering matters more. A smaller screen resistor gave a nasty tone I didn't care for. too little screen resistance resulted in some strange distortion I didn't like.
Roe wrote: ↑Mon Jul 08, 2019 7:55 am
Plate voltage is around 320 and screens a little lower with a 470R shared screen resistor in series with 100R individual resistors
How was the bass response affected by the change in screen voltage?
Not much since I replace a choke with a 420R resistance with a 470R resistor. the CLC filter and increased filtering matters more. A smaller screen resistor gave a nasty tone I didn't care for. too little screen resistance resulted in some strange distortion I didn't like.
Roe,
I imagine your amp to be very quiet with the CLC in front of B+1. Very low ripple. With just the traditional setup of 16uF for the reservoir cap, the ripple is 10V pk-pk. 32uF cuts that down to 5V pk-pk. Did you notice a change in feel or dynamics going to the CLC supply?
That's an interesting observation about the smaller screen dropping resistor giving bad tone.
