OTL Headphone Amp project.

[Simon Barden]

You could try selling the 43k resistors in packets of say 10 & 50 with a hefty mark-up on eBay. Though that might be a good place to look for some in case someone has done the same thing.

[DrNomis_44]

You could try selling the 43k resistors in packets of say 10 & 50 with a hefty mark-up on eBay. Though that might be a good place to look for some in case someone has done the same thing.

Yeah, I guess I could, it'd be a good way to recoup my money.

[Simon Barden]

Have you tried looking elsewhere for smaller quantities?

[DrNomis_44]

Have you tried looking elsewhere for smaller quantities?

Not really, because 43k isn't one of the common resistor values.

[Simon Barden]

Could you make one up from a couple of standard ones in series or in parallel?

[DrNomis_44]

Could you make one up from a couple of standard ones in series or in parallel?

Two 82k resistors in parallel would give about 41k, there are three resistors in the 6N1-P OTL Headphone circuit that are critical to it's operation, the 43k anode resistor for the triode gain stage, the 200 Ohm resistor, and the 220 Ohm resistor in the push-pull output stage, these three resistors set the balancing of the push-pull output stage for minimum distortion, and the ability to drive impedances down to 32 Ohms, if the balance is out the output gets badly distorted when driving headphones as low as 32 Ohms.


The resistor values used in the 6N1-P OTL Headphone amp circuit optimise it so that 6N1-P Valves operate properly, but the HT supply needs to be at least +348V DC for them to perform at their best.

And since the push-pull output stage is directly-coupled to the 43k resistor, the 43k resistor effectively sets the bias point of the push-pull output stage.

https://headwizememorial.wordpress.c...one-amplifier/


Remember, this is a piece of Valve HiFi gear I'm making, although there's no reason why it can't be used for guitar purposes, I reckon it'd make a great line-amp for recording purposes if you want to record an acoustic/electric guitar with piezo pickups into a Daw, you could also use it with an FX processor pedal, like the Digitech RP 50, then you'd have a good practice setup so you could practice your guitar playing and not annoy the neighbours.

[Simon Barden]

If you look at the parts list, that 43k resistor is actually given as a 42k resistor.

So either the 43k on the diagram is a typo, or the builder chose 42k over 43k as being more readily available. The "261-42K" Mouser part number no longer exists, so i have no idea what tolerance resistor they'd opted for.

Note that the OrCad diagram and the hand-drawn sketch also shows a 42k 2W resistor.

So I feel that 42k is the correct value. But you can obviously build it using 42k, measure the bias current and look at the output waveforms and decide whether you need 43k. In which case you could simply wire a 1k resistor in series with the 42k resistor.

Or if you've got the OrCad programs mentioned (they will have moved on since the 20+ years since that web site was created) you could try loading the simulation files and try with a 42k vs 43k resistor.

[DrNomis_44]

If you look at the parts list, that 43k resistor is actually given as a 42k resistor.

So either the 43k on the diagram is a typo, or the builder chose 42k over 43k as being more readily available. The "261-42K" Mouser part number no longer exists, so i have no idea what tolerance resistor they'd opted for.

Note that the OrCad diagram and the hand-drawn sketch also shows a 42k 2W resistor.

So I feel that 42k is the correct value. But you can obviously build it using 42k, measure the bias current and look at the output waveforms and decide whether you need 43k. In which case you could simply wire a 1k resistor in series with the 42k resistor.

Or if you've got the OrCad programs mentioned (they will have moved on since the 20+ years since that web site was created) you could try loading the simulation files and try with a 42k vs 43k resistor.

I have been thinking about getting myself a copy of the Orcad software, I did play around with another bit of circuit simulator software called Circuitmaker 2000 if my memory serves me right, it did occasionally produce some funny results though, I wonder if you can still get Orcad.

[Simon Barden]

Yes, I found their web site.

https://www.orcad.com/

But it now looks very professional, so could be expensive.

But their are free trial versions, which could be enough to see if it will load the 21-year-old configuration file. The format may well no longer be supported,

Otherwise you’d have to recreate the design from scratch.

[DrNomis_44]

If you look at the parts list, that 43k resistor is actually given as a 42k resistor.

So either the 43k on the diagram is a typo, or the builder chose 42k over 43k as being more readily available. The "261-42K" Mouser part number no longer exists, so i have no idea what tolerance resistor they'd opted for.

Note that the OrCad diagram and the hand-drawn sketch also shows a 42k 2W resistor.

So I feel that 42k is the correct value. But you can obviously build it using 42k, measure the bias current and look at the output waveforms and decide whether you need 43k. In which case you could simply wire a 1k resistor in series with the 42k resistor.

Or if you've got the OrCad programs mentioned (they will have moved on since the 20+ years since that web site was created) you could try loading the simulation files and try with a 42k vs 43k resistor.

I just remembered that I've got a whole heap of 1W/1% Metal Film resistors, and I may be able to find four 82k resistors amongst them, so, two of them in parallel will give me a 41k/1% 2W Metal Film resistor.

As luck would have it, I can only find three 82k/1W/1% Metal Film resistors amongst them, and four are needed.