The bridge is dissipating a lot because you’re shorting current to ground to achieve your voltage. It’s probably taxing the transformer winding too. If you need to drop a bit of voltage use a series resistor.
you’re right… the secondary is running super hot, i didn’t want to use a resistor because of the huge variability between the tubes. I might have to get a different filament transformer that has enough headroom for regulator dropout, then I can scrap this whole section and replace it with a normal regulator
Yeah but my REW software for FR sweep measurements is not on my scope, it's on my PC... which only has a line input
You do not need REW to do simple frequency response measurements. Some math required plus familiarity with your scope.
Check out Rod Coleman's dht filament regulators.
Check out Rod Coleman's dht filament regulators.
I think if you're going to start digging into testing, a USB scope would be worth the investment. A Picoscope 2204 is only $130 and comes with very good software. It has a built-in generator and adequate bandwidth for your purposes. It will enable you to do a lot things--check square wave response, which is often more revealing than a simple frequency sweep, test for feedback levels, power output, and so on.
Yes. Rod Coleman regs would be a good investment. Another option is member V4lve Lover's 6V-in-6V-out board, which can be adjusted down to 5V. I use one of these in my Aikido preamp and it's a nice board, very handy if you only have a 6V tap available.
Do not think that you will get snappy midrange and detailed hights with 300b.
Even if you make the amp to roll at 60kHz, I doubt that will make any audible difference.
The kicker is that I spent decent money buying a Hammond 300BX with all the taps necessary…… for an AC heated amp…
I was getting hum out of the speakers with NO B+! and checked to make sure it wasn’t transformer coupling.. Took a few days to accept the fact that i’d have to use a separate filament transformer and go DC
I’m aware of those filament power modules such as the coleman but I’d still like to give it my own go. I wouldn’t be in this hobby if I wasn’t learning things the hard way and dismantling my assumptions and learning here my blind spots were. I may sure enough arrive to the understanding as to why so many people go with the rod colman regulator or just get a tube lab SE board but to me that sort it takes away the fun/adventure.
As for measuring FR between stages using a standalone scope or getting a USB scope, i’m not opposed to getting something like an Analog Discovery or picoscope, but i’m not yet convinced that I can’t just make some sort of robust, low impedance adaptor to convert HV to line level… I think i’ll just make “half” a tube pre, If I destroy my audio interface/computer, then so be it, i’ll get a USB scope 🙂
I was getting hum out of the speakers with NO B+! and checked to make sure it wasn’t transformer coupling.. Took a few days to accept the fact that i’d have to use a separate filament transformer and go DC
I’m aware of those filament power modules such as the coleman but I’d still like to give it my own go. I wouldn’t be in this hobby if I wasn’t learning things the hard way and dismantling my assumptions and learning here my blind spots were. I may sure enough arrive to the understanding as to why so many people go with the rod colman regulator or just get a tube lab SE board but to me that sort it takes away the fun/adventure.
As for measuring FR between stages using a standalone scope or getting a USB scope, i’m not opposed to getting something like an Analog Discovery or picoscope, but i’m not yet convinced that I can’t just make some sort of robust, low impedance adaptor to convert HV to line level… I think i’ll just make “half” a tube pre, If I destroy my audio interface/computer, then so be it, i’ll get a USB scope 🙂
That may very well be, and that’s ok 🙂 I really do like the presentation this amp provides otherwise, it’s just the top end detail that i’m missing
You will always get hum with AC filaments when the B+ is off. ;-) With the B+ on, the hum will disappear (mostly) with a balancing pot. I don't know the technical explanation but it's true.
With a USB scope and a laptop on battery power (so the laptop isn't grounded to mains), there is very little chance of damaging anything when taking standard measurements. For a standard PC, you should google this to be sure, but as long as the equipment you're measuring is floated you're okay. You don't want any ground differential between the computer and the piece you're testing. Measuring at the grid of the tube, after the coupling cap, there's no DC on the probes. For AC measurements of more than 20VDC, use the 10X setting on the probe. Leave your sound card out of it, the Picoscope has a built-in generator.
For high voltage measurements, like balancing a phase-splitter at the plates for lowest distortion, you use a special differential probe.
Thousands of people use Picoscopes to measure audio equipment, it's not a big deal. ;-)
I like the software of the AD but the Picoscope looks more more like complete product with BNC connectors and half the price. I guess now it’s a matter of can I build an adaptor sooner than I can order in a pico-scope… I suppose i’ll be revising the filament supply this weekend anyways which should keep me pretty busy.
Not my experience at all, I have been building 300B based amps for over 25 years and it's quite possible to get that kind of performance out of the 300B as well as quite a few other DHTs. My experience is mostly with JJ300B, EML300B, and early 2000s SV300B, and WE300B. (Also some 1950s 300B) I use fixed bias in the output stage and either IT or hybrid drivers with tubes like the D3A.
I'm still puzzled by why you feel you need an "adaptor." With standard 10X probe you should have no problem measuring the frequency response of your driver stage at the 300B grid.
The Picoscope software is very good and easy to use. There's also a handy 3rd party app for Windows that compares input to output and creates a response plot that can include phase shifts and gain margins. It's more for testing feedback networks but is great for running FR plots.
https://www.picotech.com/library/picoapp/frequency-response-analyzer-with-bode-plots
In a scenario where I want to measure it with my PC sound card using REW…
I wouldn’t need to do this with with a USB scope.
I ran it without C4 and C5 for years. I added those about a year ago to attempt to improve top-end but to no avail.
The HF feedback added by C4 and C5 will actually further reduce the gain where you need it most. 🙂 The source impedance of your driver stage is extremely high (at least a few hundred K ohms). A high transconductance video pentode with a 10K plate load resistor should give you the swing and FR that you need if you want to stay with that topology, and with much lower source impedance.