LOW VOLTAGE 845 SE

zygadr · « **on:** 18 Aug 2009, 06:43 am »

Hi everyone, I have just bread boarded a low voltage (450 plate volts) 845 S.E. amplifier as such :

600ohm : 10k input transformer connected to a paralelled 6SL7,cap coupled to an 845, cap coupled to another 845 and then connected to a 5K output transformer.

The sound is quite amazing.......weighty and transparent

- totally different sound to a previously built and owned 300b SE which sounded veiled in comparison(Reichart circuit, Sovtek 300b's).

Anyone done the same or like to comment on this?

Niteshade · « **Reply #1 on:** 18 Aug 2009, 10:04 am »

Very interesting!

What is your power output?

I was thinking along similar lines:

1. Use around 1200v and a 15K primary output transformer.
2. Use the 572B tube

Transformer coupling is a neat idea.

zygadr · « **Reply #2 on:** 19 Aug 2009, 07:14 am »

Power output would theoretically be around 5-6 watts.

The use of the input transformer as per the SAKUMA style amplifiers is absolutely BRILLIANT!!!....even cheap transformers sound better than most valve input stages!

I could go higher powered, but it would require some major component changes.........who knows........might try it sooner than later......all depends on whether it will drive my large diy NXT panels

Niteshade · « **Reply #3 on:** 19 Aug 2009, 12:54 pm »

Graphite plate tubes have great potential. I want to see around 50 watts out of one of these guys (572B). That will take some drive- maybe 5 watts. However, since most large SE output trannies go to 25-35 watts, I'll settle for 35 with lots of headroom. Some gentle, quiet forced air cooling will be in order too.

It's a great tube for a great price!

smbrown · « **Reply #4 on:** 19 Aug 2009, 01:58 pm »

The only issue with input transformers is they can require a very low driving impedance. Otherwise I really like the idea of breaking ground loops with them. Since you have 2x 845, have you considered going the full Sakuma and doing a PP? Have you seen Lynn Olsen's work in this area?

JoshK · « **Reply #5 on:** 19 Aug 2009, 05:57 pm »

Quote from: Niteshade on 19 Aug 2009, 12:54 pm

Graphite plate tubes have great potential. I want to see around 50 watts out of one of these guys (572B). That will take some drive- maybe 5 watts. However, since most large SE output trannies go to 25-35 watts, I'll settle for 35 with lots of headroom. Some gentle, quiet forced air cooling will be in order too.

It's a great tube for a great price!

You could use a Source Follower like what is done in the power drive circuitry.

I've also seen a 833 SE amp where the B+ was ~600v but the power output was up to 75w (stated). It was basically a zero biased triode using a pentode (EL34 (iirc)) as a CF to drive the output. SF might be cheaper as just as effective.

zygadr · « **Reply #6 on:** 20 Aug 2009, 08:12 am »

Yes, I have read all the literature on the Sakuma and the ''il monstro'' article''.

Last night I managed to insert a Hammond 124D interstage transformer in place of the coupling cap between the 845''s.....WOW

......big difference......a lot louder and more authorative.

I'm trying to stick to Sakuma's shematic as close as possible. The only difference is parts choice and the use of a 6SL7 after the input transformer.

My only problem is that it humms quite a bit.......the hum pots are at maximum and the slightest turn backwards causes the hum to go through the roof.

I'm using A.C. on the filaments but did a quick rough D.C last night and there was not a great improvement(some buzzing or oscillation was evident.).

Any ideas on the hum pot maxed out anomaly??

JoshK · « **Reply #7 on:** 20 Aug 2009, 03:46 pm »

Are you floating your 845 filament supply? No unintentional groundings?

Are you using a seperate filament transformer or the PTx? How much ripple do you have (careful measuring it)?

zygadr · « **Reply #8 on:** 21 Aug 2009, 06:58 am »

JoshK, discovered that most of the hum was coming from my power supply. In my hurry to knock up a quick working prototype, I made an error in wiring which left out a capacitor and the choke out of the circuit!!

Anyhow, hum has reduced by about 80% if not more. But, there is still hum.
I am using separate toroidal trannies on the 845 heaters. Can't see any grounding mistakes, but the wires are all over the place at the moment, so who knows?

I still think that I need to go CLC or CRC on the filaments with D.C. instead of A.C.

I have some reasonbly sized 1H chokes that might(

) do the trick without melting themselves down

smbrown · « **Reply #9 on:** 21 Aug 2009, 01:51 pm »

In my experience, the higher the heater voltage on DHT, the harder it is to get hum free. I can get 300b's quiet w/AC but go to 6.3v and higher, much harder. I remember when the SV 811 (6.3v) was being produced, many builders said it had to have DC, and I agree. The 845 might do ok w/AC in PP where you can cancel much hum in the output (common mode).

zygadr · « **Reply #10 on:** 22 Aug 2009, 11:06 am »

smbrown, that's interesting to note. So, in S.E. mode, D.C. is recommended ?

Actually the A.C. voltage at the moment is around 9.3 volts most of the time. So it's hard to tell what's going on until I up the voltage. D.C is the only quick way I can do this.

zygadr · « **Reply #11 on:** 23 Aug 2009, 11:26 am »

Are Australian made, 5K, LANCROFT.

It's easy to blame the inexpensive Hammond 124d interstage, but I remember someone at Audioasylum abandoning his project with the same output transformers due to a ''lack of bass''.

Anyone care to comment on why this could be so?

zygadr · « **Reply #12 on:** 24 Aug 2009, 07:28 am »

The Hammond 124D interstage was taken out of the circuit and replaced with a 1.0mf oil cap.
Bass has returned and surprise!, surprise!........turning the hum pots have COMPLETELY removed the HUM

........don't ask me how....it just has.

Next step will be the removal of the input transformer, and reconfigure the 6SL7 input stage to a series connected (both halves of the valve)two stage input which is cap coupled to the first 845 as per Bonavolta's web site. Hope this will work with sufficient gain as intended.

JoshK · « **Reply #13 on:** 25 Aug 2009, 01:47 am »

How much current are the Lancrofts gapped for? What are you running them at? Do you happen to know how much inductance they have, particularly at low frequencies?

There are a lot of reasons for lack of bass, weak OPTs are one of them. But I wouldn't be too quick to blame the OPTs.

JP78 · « **Reply #14 on:** 25 Aug 2009, 02:12 am »

i understand this varies from the discussion a bit, but i have a bel canto integrated that runs a single 845 output per channel and i'm getting about 37wpc in class a set operation. fwiw, the bel canto techs told me it is rated for operation down to 2 ohm nominal.

zygadr · « **Reply #15 on:** 25 Aug 2009, 06:56 am »

Josh, as the original manufacturer has passed on the business to someone else, it's very hard to know what the transformer's specifications are.They were specifically designed for 300B's not 845's, so that could be a problem.

It's a pity about the interstages as they produced a very clean mid/highs, just not much bass(but plenty of gain - which is what I need.)so it looks like I'll have to invest in some real world interstages , as cap coupling brought the bass back.

Can anyone recommend any 5k : 10k interstages that don't cost a fortune?

JoshK · « **Reply #16 on:** 25 Aug 2009, 12:29 pm »

What's a fortune? Lundahl makes interstages at different price points, and is said to be quite good.

You might dig around on the net, particularly diyaudio's tube forum and see if you can find comments on the 124. It may be that you are wiring it suboptimally. I know you have to be careful about parasitic capacitance to ground with interstages.

I'd think it'd be worth investigating the root cause of the lack of bass due to the 124. Maybe it just isn't enough inductance, that is possible too.

300B OPTs are typically gapped for 90-100mA. How much current are you running in your amp? Too much and you are saturating the core which would lead to lack of bass.

zygadr · « **Reply #17 on:** 26 Aug 2009, 03:52 am »

I'm running the 845's at about 70ma....so that part seems o.k.

Have reconfigured the paralell 6SL7 to a cascade stage and the sound is now quite respectable.
The input transformer prior to this 6SL7 stage will now need to be upgraded.

I can obtain a 200-600ohm: 50k good quality microphone input transformer for $100 AUST......made by O.E.P in England.
Otherwise it may have to be a lundahl or something vintage from EBAY?

JoshK · « **Reply #18 on:** 26 Aug 2009, 12:38 pm »

Better have one heck of a preamp (mini amp). Driving 6SL7s through a 1:10 stepup transformer is no small feat. Its a recipe for no highs due to miller capacitance and slewing.

zygadr · « **Reply #19 on:** 27 Aug 2009, 03:20 am »

Josh, maybe so, I have the amp connected to a single 12'' twin cone loudspeaker from the early sixties (Australian Magnavox) so highs have not been expected or listened for all that much

As an update, I have now changed some operating points on the cascade 6SL7 input stage after some hair raising voltage discrepancies and now have more gain than I know what to do with!

........so, I can toss the input transformer if It seems the ''tinklies'' have dissapeared