25W Single Ended Hammond 193V Choke Loaded 2SK180 L'Amp

[adrianrw]

Thanks @pinholer and @Ben Mah. The shoulder washer has a very shallow depth, maybe 0.2mm. I have added a bit of shrink tube, now the RCA input won't make contact with the case.

I have also implemented the grounding as advised. Below are some pictures. I'm busy the next couple of days, but hopefully I get around on Sunday to start testing the amp by powering it stage by stage up.

 

[adrianrw]

@Ben Mah the amplifier works🎉 A couple of notes for the power up:

• I followed you power up steps, thank you for the details
• I measured 1.1Ohm for the choke
• With no load attached to the V- power supply I measure -45V
• Testing the amplifier board with two multimeter say A & B
  • A: T.P. SIT S Choke2 & T.P. V- Choke-1
  • B: T.P. Choke2 SIT S & T.P. GND
• With full potentiometer fully counter clockwise I measure at B 43.97V
• With Iq set to 2.5A (i.e. potentiometer set so that A measures 2.75V) I measure at B 36.8V
• I currently measure 65°C at the heat sink (maybe after 15min)

I've attached test speakers and used my chord mojo as input and it sounds good, no audible issue.

I'll now assemble the second amplifier (left channel) and I'll have them ready to demonstrate at the Burning Amp 2024.

Thank you very much again @Ben Mah! I'll keep posting pictures and comments with my progress and impressions.

 

[Vunce]

What are the dimensions of your heatsink? 65°C after only 15 mins is hot, what is the temp after an hour? Probably should lower bias current or use a helper fan.

 

[Ben Mah]

Hi Adrian,

It's great that you got it up and running successfully on the first power-up!

65 degrees C is a bit hot. What size is your heatsink and how did you measure the temperature?

My amps have the Modushop 300mm x 210mm x 40mm heat sinks and at 2.5A, they are only moderately hot - much less than 55 degrees C. I have run them at 3.0A and they did not get to 55 degrees. They didn't even get to Blimey hot.

 

[adrianrw]

I am using UMS 40mm Heatsinks from the diyaudiostore. Each is 165mm x 300mm x 40mm. I am using an inexpensive infrared thermometer gun and point it at the center of the heat sink where the THF51S is mounted.

Also, at louder volumes I do get audible distortion. I tried to buffer the signal with a Pass Korg Nutube B1 pre-amplifier but the distortion stays. Low volumes seem to be good. I thought I'll measure next with a signal generator and a 8 ohm dummy load and a picoscope if the distortion is visible.

 

[Ben Mah]

How long can you keep your hand on the heat sink? Do you have a contact thermometer? I use a cooking thermometer to measure heat sink temperature.

The amplifier is a follower so it does not have voltage gain, so a high voltage output preamp is needed to feed the amplifier. Unfortunately the Korg Nutube B1 is not a good match as it has low voltage output and high distortion. Nelson's measurements show maximum 5V output at about 7 or 8% THD.

A very reasonably priced and great sounding circuit that is compatible is Nelson's DIY Front End 2022. Currently I use that as my preamp.

 

[Ben Mah]

Further to my suggestion of the DIY Front End 2022, it does not have to be configured as a preamp. It can be installed in the amplifier chassis and powered by the amplifier power supply. It would behave as the first stage, the voltage gain stage, of the amplifier. The RCA input would go into the DIY FE, and the DIY FE output would go into the amplifier board.

 

[adrianrw]

thanks @Ben Mah. I will try to use a contact based thermometer.

Regarding the FE 2022, I do have two essentials kits that I have not assembled yet. If I make it as a permanent buffer in the amplifier then would I have to connect it to the -37V (V+ of PS to V+ of FE and V- of PS to V- of FE) and short R9 as Nelson Pass describes in his article?

Alternatively, I did order a Meanwell 48V/30W power supply, for initial testing, but I assume that will perform not as well as the power amp power supply. What power supply did you use for the external FE2022?

 

[Vunce]

IR temp guns do not give accurate readings pointed at shiny metal. Aim at the heatsink that is outside of the chassis and is anodized black. The inside is plain ‘shiny’ aluminum.
Or, as Ben suggested, use a contact thermometer or probe for your DMM.

 

[Ben Mah]

Yes, for a negative voltage power supply, R9 is shorted and the V+ pad becomes Ground (for signal input and signal output of DIY FE 2022, and amplifier power supply ground), V- of amplifier power supply connects to the V- pad.

If you mount the FE board inside the amplifier you do not need another power supply, just use the existing amplifier power supply.

My DIY Front End 2022 preamp build here: https://www.diyaudio.com/community/threads/diy-front-end-2022.394339/post-7312224,

and power supply details here: https://www.diyaudio.com/community/threads/diy-front-end-2022.394339/post-7313031

Heat sink temperature: how hot is it to touch? How long can you keep your hand or finger on it?

 

[adrianrw]

@Ben Mah i think I figured out one issue, I’ve put the LEDs in reverse polarity, i.e cathode (negative) to the square pad as in the picture below. But I think in your pcb design square pads are for the anode, correct?

 

[Nelson Pass]

IR temp guns do not give accurate readings pointed at shiny metal. Aim at the heatsink that is outside of the chassis and is anodized black. The inside is plain ‘shiny’ aluminum.

Or, as Ben suggested, use a contact thermometer or probe for your DMM.

Use a black felt pen to color the surface of the spot you are measuring. Works like glue...

 

[Ben Mah]

I just looked at the Kicad schematic for the amplifier pcb and it looks like I put the LED symbol in backwards. So if it is the LED in the bias supply on the amplifier pcb board that is not lighting up, it is my fault for the reversed LED. I didn't catch the mistake because when I install LEDs in a board, I make sure that the anode (long lead) is at the positive side of the circuit.

 

[Ben Mah]

After looking at the circuit some more and trying to remember the design, I think that the pcb is correct and there is no error in the LED polarity. So ignore post #973.

Adrian, which LED are you having the issue with?

I just measured the voltage and polarity of the LED in my amplifier pcb and the LED pad (square for cathode) on the pcb is correct.

I examined the power supply pcb and the cathode pad on that is also correct.

 

[ItsAllInMyHead]

I just measured the voltage and polarity of the LED in my amplifier pcb and the LED pad (square for cathode) on the pcb is correct.

I ignore square pads noting polarity for this exact reason ... I get easily confused.

For caps (as an example) many people use the square pad for the positive lead / long lead, which I understand to be standard.

As I understand it for LEDs, the standard convention (as you've done), is to use the square pad for the negative (cathode) / short lead...

One of my favorite PCBs uses the long-lead / square method for LEDs ... so... it seems it varies across designers.

I just ignore it and measure them like you've noted. DMM on diode mode... make LED light up (with my boxes of 100s of cheapies I've gotten duds). Stick it in with long lead to positive side of circuit using the schematic symbols.

Heck, I just did a "Google" search for "Square Pad for LED" just for fun (and because I was curious about what's normal for LEDs) ... and look what came up... So, no wonder people get confused... Always trust the first search answer on the interwebs. Is it the long lead or the cathode? Is it the flat side or the anode.


Cheers!

 

[Ben Mah]

The long lead on the LEDs that I have is the anode. That is the standard convention. However there may be some LEDs out there that are reversed.

Don't believe everything you see on the internet.

I also check my LEDs with a meter in diode or LED mode before installation. I have damaged LEDs by keeping them connected to the meter for too long of a period of time. So now I just connect one lead to the meter probe and tap the other lead with the other probe to see if the LED lights up. To do this check, connect the short lead of the LED to the black probe and touch the red probe to the long lead. If the LED lights up, the long lead is the anode and the short lead is the cathode (red probe = LED anode, black probe = LED cathode).

 

[ItsAllInMyHead]

Don't believe everything you see on the internet.

Exactly... that's why all the laughing emojis. The fact that literally the first search result I pulled goes against all conventions... was a bit funny (to me). To a beginner that has no idea... it could be frustrating.

Edited to add - we both may be being a bit kind... I'll be so bold as to say... it's just factually incorrect re: relative leg lengths and the associated nomenclature.

 

[adrianrw]

Thanks for all the inputs. I’ll wrap up the second mono block and the get into measurements again.

@Ben Mah i am building an FE2022 with an SMPS, I’ll leave room for a linear power supply for later. What gain R3/R1 do you recommend? I have a 100k ohm Alps pot.

Also out of curiosity, why does everyone use a potentiometer to modulate the input signal instead of replacing R3 with a potentiometer to control the volume?

 

[Ben Mah]

Go with the values on the schematic, which is for 10x gain. For the power supply go with 50VDC as Nelson suggested.

R3 is within the circuitry of the FE. Replacing it with a variable resistance may have unintended consequences, which is beyond my comprehension.

By the way, did you get an accurate measurement of the heat sink temperature, and what was the issue with the LED?

 

[Ben Mah]

Just a thought about the DIY FE 2022, uprating capacitors C4 and C5 from 25V to 35V would give you a bit of headroom and flexibility with the power supply voltage.