• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

My KT88 Williamson Amp Build

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Really nice job!

I'd like to see the schematic for the delay circuit, please.

Here it is. Note that on the relay part number specified the coils are actually DC polarized, so you have to make sure the polarity is right or they won't engage. Part numbers on the schematic are from Digikey.

An externally hosted image should be here but it was not working when we last tested it.
 
Hummm, I can't see where the relay contacts integrate with the other drawings. Do you have that illustrated? Looks like a 50 second delay with those values?

I've never done a time delay circuit with solid state rectification but I would be tempted to put the contacts on the transformer secondaries; my thinking being that to do so would provide a softer start. What is your thinking?

Thanks for posting the drawing.
 
Hummm, I can't see where the relay contacts integrate with the other drawings. Do you have that illustrated? Looks like a 50 second delay with those values?

I've never done a time delay circuit with solid state rectification but I would be tempted to put the contacts on the transformer secondaries; my thinking being that to do so would provide a softer start. What is your thinking?

Thanks for posting the drawing.

The relay is located between the secondary diodes and the 10 Ω resistor on the high voltage supply. Do not delay the bias supply. 50 seconds is about what my amp takes to power up.

I have only provided the main modules of the amps. Exact details of the design I would prefer not to publish, but there is enough detail to allow the knowledgeable builder to put together his or her own working amp with what is published.
 
Thanks for sharing.

Not to fixate on such an inconsequential detail, but B+ delay is on my mind at the moment for a current project. I notice that placing the contacts on the DC side of the circuit as you have done is most common, yet it appears to me that the contact ratings on relays usually favor AC over DC (higher amperage for AC at the same voltage rating). Is there an over-riding factor I'm failing to consider?
 
Novice/Noob question

Hi there Loren 42,

I'm just starting to learn about building DIY valve amps. In particular KT88 designs. Thank you for detailing and posting this expertise/project.

I can see this thread is a bit old, but it's still open.

I'm studying the schematic and the photo of the T-Rex. I can kind of follow/understand it. But I'm stuck, the schematic appears to show 6 valves per channel, but the in the photo of T-Rex it only seems like there's 4 valves per channel :confused: Is the photo and the schematic the same circuit/design? Sorry if this is really obvious. :eek:

Would you be able to describe/summarise/overview the basic circuit in the schematic?

Thank you
 
Hi there Loren 42,

I'm just starting to learn about building DIY valve amps. In particular KT88 designs. Thank you for detailing and posting this expertise/project.

I can see this thread is a bit old, but it's still open.

I'm studying the schematic and the photo of the T-Rex. I can kind of follow/understand it. But I'm stuck, the schematic appears to show 6 valves per channel, but the in the photo of T-Rex it only seems like there's 4 valves per channel :confused: Is the photo and the schematic the same circuit/design? Sorry if this is really obvious. :eek:

Would you be able to describe/summarise/overview the basic circuit in the schematic?

Thank you

Well, the question of the number of tubes has been answered, but I can give a quick circuit description.

V1a is simply a buffer stage with gain. It drives the second stage, V1b, which splits the signal into two, but in opposite (180°) phase.

The splitter is called a concertina phase splitter. The input signal is split and tapped off at the anode and cathode of V1b. The phase coming out of the anode is 180°*out of phase from the input, while the cathode is in phase with the input.

The critical thing here is that both the upper and lower resistors be as closely matched as possible.

I bought many of these resistors and hand matched them to get as close as possible. This way both output signals will have the same amplitude.

Now, V1a amplifies a 2 VAC peak to peak input signal to about 30 VAC peak to peak.

V1b actually reduces the gain to about 27 VAC peak to peak, or about 90% of the input signal. That is because of the cathode follower geometry. You always get slightly less than unity gain from the cathode. Since both outputs are now matched, the voltage drive is slightly reduced.

This leads us to V2, which takes our symmetrically split signal and amplifies the signal to something that will drive the KT88s to full power.

Notice that V2a and V2b both have carefully matched plate resistors of 47K.

R22 is the feedback resistor. You will need this to keep the circuit from oscillating and if memory serves, you may have to reverse the plate and screen connections on the transformers if the amp oscillates. I think the Edcor output transformer does not color code the phase relationship of the windings, so it is a trial or error method. My first attempt was an error and resulted in high frequency oscillation, but swapping the leads fixed it.

You need an oscilloscope and a dummy load to check this as the oscillation is well above hearing. Your only clue is that the tubes will be running hot as they produce full power into the load.
 
Is there a schmatic for the stereo divider board? Pretty please with a cherry on top? :)

The driver board contains everything from R2 through C3 and C4 on the schematic in my first post (see page 1 of this thread).

After C3 and C4 the parts are hand wired on the chassis at the tube sockets.

Here is the PCB layout I created:

An externally hosted image should be here but it was not working when we last tested it.


Note that the filaments are wired directly to the sockets, while the PC board is actually elevated above the sockets with a metal ground shield between the PCB and the filament wires. I attached bus wires to all the relevant pins on the octal sockets that are soldered to the pads on the PCB. This may have been a bit excessive, but that was how I did it.

R1 (shown on the schematic as a 100K pot) on the PC board is/was actually a 10K resistor to ground.

I have since added a 10K pot for the input volume control and a pair of Cinemag balance line transformer CMLI-15/15B

You can see how everything is wired up in the following PDF:

PCB Interconnects

The balanced line inputs allowed me to delete the safety ground diodes and connect everything (signal and earth grounds) directly to the chassis at a single point (star ground). The balance line transformers will also break any ground loops from the rest of the equipment that drives the amp.
 
Last edited:
Thank you. The balanced line inputs are added value to the project. The whole project appears to be very well executed & documented. The only other feature I might wish to add, would be meters for biasing - it they were applicable.

I have provisions for test points accessable from the top side of the amp. You can see them on the PCB Interconnects PDF file and if you look close in the photographs they are behind the KT88s on the 1/4" plate near the trim pots.

The pots and test points are mechanically recessed, so they are not easy to see, but easy to get to.

Voltage should be .540 VDC each or so. I have to double check that.

I have found that once you install new tubes you have to set them twice, then they seem to be very stable from that point onward. Not worth installing permanent meters for in my opinion when I have a DVM handy.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.