Thanks for commenting. I've been happy with my Goldpoint attenuators and would purchase from them again. Their website also has some good resource material for DYI'ers. Cheers.
Been using a 5670 tube buffer PCB kit now for a year or two, and I absolutely love the thing! I wanna do a similar project, just without the bread boards. I'm gonna grab an input relay/remote volume kit, soft on/off board and a PSU in a similar enclosure. Super nice project here, great job!
@@MikeFreda I discovered a message board post going back where someone had done all the leg work, testing almost every tube of this type out there, and concluded the JAN GE milspec tubes had the most pleasing sound overall, mainly in the bass dept. I had my reservations with the dinky kit I got as well, but low and behold I prefer it in my system than without, which ultimately lead me to researching how I can make it into a full legit linestage and not some PCB sandwitched between pieces of acrylic.
This is great, thanks. I built all solid state gear. Wanted a tube pre but the gain is usually to high and output impedance didnt match amp. Might give this a go.
I’m assuming you are inquiring about the CCS board. I have another video about installing CCS on my 6DN7 amp. I go into detail about this board, how it works and also have a downloadable worksheet. Cheers.
Interesting, you will have to get a bit creative to adapt 6SN7 to a OTL headphone amp. You may need another tube. Love to hear how you make out. Cheers.
Not really. Do not recommend. A 12AX7 has too much gain (mu of 100) for this application and the output impedance is also different. You would better luck with a 12AU7 or a 12AY7 but you would need to select different output transformers due to the higher output impedance of these tubes. Sort answer, no. It’s the wrong tube for the job. Probably be noisy as well. Cheers.
Can I make a suggestion? As a recording engineer and musician, who wants to build one of these myself May I suggest that you have dual functionality one for recording and one for monitoring through your daw digital audio workstation As I believe the tubes need to be somewhere else in the circuit to do these two jobs separately, which can be easily fixed as long as the box is set up right from the beginning So then you want to monitor your mix when you’re not recording through the tubes Generally, for mastering purposes
@@MikeFreda I don’t think so I am definitely not a tube expert myself. I have some audio engineering skills, but of the electrical nature none Although I am very interested in the subject But I can point to do a video where somebody mentioned that concept that I am talking about
Excellent... Not to easy to find that high quality part in my local market (even in my country). Just one thing, why that headphone jack sets in the back side?
Yes, this preamp design does require quality components to make it sound good. The headphone Jack is an afterthought. I wasn’t sure if it would work. It was just easy to install it right beside the output…. But yes, it would easier to use if it was on the front. I would use shielded cables. That being said, I don’t really use this as a headphone amp (I already have a dedicated one) so it’s just there for proof of concept and to see if it could drive high impedance headphones, and it can. Cheers.
Hello, thanks. I'm fortunate to have a really good electronic store near me and I'm able to get a lot of my supplies from. This is some of the hook up wire I use: leeselectronic.com/qc/product/224902-hook-up-wire-22awg-solid-ul1007-red-50ft-spool.html They also sell 2 wire, shielded wire from spools. leeselectronic.com/qc/product/23814-high-voltage-2x22awg-w-ground-ft4-jacked.html I've found similar wire at Mouser but they typically have spools and it can get pricey, cheers.
The input impedance of the 5670 is around 100K. I mostly use a turntable (with a phono stage) and a NAD CD player which they typically like to see 50K or higher. Their output levels seem to pair well with it this pre-amp design. You could try 10K if you like. I'm not sure what the end effect will be. Cheers
Well done, finally line stage with appropriately low gain, unlike all those ECC83 stages having way too much gain and weak drive (good for 100k but not 600r). I like use of veroboards because they allow for optimal layout with either small pins or tagboards. Tube is biased very hot. Didn't do calculations for 600r load like headphones but they might do fine with lower dissipation when 15k:600 OT's are used, i guess such biasing was used for the sound. I'm very interested in seeing turret or vero board layout for two stages of class a push pull circuit with tubes mounted horizontally. If anyone ever made mic or phono preamp of this type please let me know, layout for vertically mounted tubes would also help because it is not that different in terms of layout.
Thanks for commenting. Max dissipation for a 5670 is 16.5ma. I'm running around 8ma per section which is 1/2 of max and aligns with the 150v curve. As an input tube you would go way less than that. It's a balance of gain and tube longevity. Cheers!
@@MikeFreda Yes, i should look at suggested biasing instead of max rating. Btw, 5670 was often used as vari gm tube for compressors with mixed results because it was not really made for this application like 6386 and other were. 6CG7 is another interesting tube for line stages very similar to 6SN7 in 9 pin bottle, others like 6N6, 12BH7 and similar small signal drivers would also do fine where gain should be very low. Is CCS really needed in class a where current is pretty much constant? Anyway, maybe try really high quality transformers like Jensen with 80% Ni cores, Lundahl has their own special cores too, although low mid distortion is different from Fe/Ni cores.
@@Signal_Glow No, you don't need the CCS, you could replace it with a plate choke. I would suggest something with the appropriate Ma rating and a large inductance (i.e.40H or more). The larger, the better... but they will get costly. You will have to know what the DCR rating is and recalculate the B+ voltage to give you ~150V at the plate. You could also change the topology from Parallel Feed to a Series Feed and use normal OT but you will sacrifice some performance and you may lower the noise floor a bit as well. The CCS does a great job of PSRR. At the end of the day, the CCS will be your most cost effective option and will give you the best performance. Cheers
Nice work. I'm wondering if this would sound better than my Wavelength Sine Preamp? It's an early version from 1994-95 using 6SN7's, Jensen input transformers, Penny&Giles volume pot and non-regulated power supply with 6X5 rectifier. There are two Hovland coupling caps in each channel. I've contemplated replacing the Hovland caps with Mundorf Silver/Oil because swapping them out in the Cardinal 300B SET amps made quite a difference. Was also wondering if a stepped attenuator would be an improvement. Maybe it would be better to build something from scratch? Reply
Thanks for commenting. It would be very hard to say which is "better" . I think they would sound different but better, not sure. Upgrading caps and a better volume control can add come clarity. 6SN7's have a reputation for sounding "tubey" with a strong midrange presence (at least that's what people feel) They were the go to preamp tube in the 80's and 90's and then just like that, people started to shy away from them. I have a 6SN7 pre as well but the topology and gain is different so it's hard to tell. This 5670 pre is much quieter. That goes a long way for me to sounding better. I like tube rectifiers in guitar amps but I feel solid state rectification in HiFi is preferred since you can easily incorporate higher value caps in the PSU, creates less heat, last longer, little to no voltage drop across the diode and its cost less. (some people also feel there is no sag either but I have no evidence in that since you would need to build similar circuits and test that theory) Cheers, Mike
@@MikeFreda Thanks for your reply. I thought Golden Dragon 6SN7s sounded great in my preamp. At the time when Gordon Rankin built the preamp, he told me he thought the Penny & Giles Pot sounded as good as any stepped attenuator he'd heard. And it was quite expensive. The caps are another story since there are many more choices available now. Several years ago, per Gordon's recommendation, I swapped out the Hovlands in my Cardinal 300B amps. The Mundorf Silver/Oil sound incredible. Though, Mundorfs for the preamp would be over $300. So I may audition other preamp topologies and tube types before upgrading. I know Gordon had moved on from 6SN7s. Probably be easier to build something from scratch than making major changes to this preamp. Thanks again.
Unfortunately, I don't. The performance of this circuit design would not be great if you removed the Constant Current Source. You could try and remove the CCS and install a ~14,000 ohm 2 watt Plate Resistor and use a 240 ohm Cathode Resistor but I'm not sure if there enough voltage swing to make that work properly. Worth a try I guess. If you want to use this tube without a CCS you basically need to rework the PSU to have higher B+ (Higher voltage PT) and better channel separation buy using an extra RC node for each channel or use some chokes. I'd recommend looking at the 5670 Tube Data Sheets for more information. I'm not sure if there is much of an appetite for me to rework this circuit without the CSS. It would be easier to build but it would be a step backwards, cheers
No, that's not what I am saying. Keep the PSU the same and just replace the CCS with a 14k 2W resistor. The tube needs 150V at the plate to operate properly. To get enough voltage swing, you need the B+ to be at least 100V more. Last cap at the PSU ~265VDC - 14K resistor 150VDC at the plate of the 5670. To be clear, I have not done this. I'm not sure how it will sound. Cheers.
Thanks for your comment. Yes, you can with these transformers and circuit. I have done this in the past but I didn’t really hear a difference into my 2A3 power amps, speaker and room setup. I suspect one would definitely hear a difference into headphones. Cheers.
For mic’s and guitar input? No, not enough gain and the input impedance are not ideal. For the output impedance, you need to see what gear you are chaining to. Could this circuit be modified, sure but you need more information. Cheers.
What do you think of Mr. Carlsons Lab Mike? Do you follow his videos, I learn a lot from you and him both. You seem like you learned electronics, Carlson seems like he eats, lives, and breathes electronics. You and Carlson are what I aspire to be. Thanks for the great vids my man.
Thanks for commenting. Yes, I know Mr Carlson and enjoy his in-depth videos as well. His channel is his life (along with his Patreon section) and he’s really focused on electronics. Great stuff. My channel is mostly about making and it’s just a hobby for me. I enjoy sharing my projects and knowledge to the world. Please let me know if there is other content you would like to see. Cheers.
This would be complex to incorporate in this preamp. There isn’t enough gain and you would also require an equalization network. What I would recommend is a separate tape preamp before this preamp. You could make a Phono style preamp and modify the RIAA LP equalization network with NAB 1/4” 7.5/15ips equalization network. That mod would be easy enough to do, just need to change some capacitor and resistor values. Cheers.
Thanks for commenting. I made the knobs from scratch. I have a metal lathe. Please have a look at my shop tour video and my other tube amp videos to see how I fabricate most of this items. Cheers.
You could probably add tone controls to this one quite easily using passive tone control circuitry, though you may find you don't need that depending on what your source material is. Every amp has its own signature and personality. Also will depend how you match it to your power amp. You could put an equalizer between this amp and your power amp if you need to compensate for room errors.
You can but with a single tube design it is a bit harder and you will probably lose some gain. A different topology would lend itself better to tone control. Cheers
I have a great integrated but if I were to try a pre I would think a great preamp is the way to go. What is the best? I am in Europe and have no workshop to do the handwork needed. I need to learn more.
Thanks for commenting. I believe having separate components is slightly better mostly due to the power supplies being separate. One issue to overcome is injecting noise into the system due to ground loops and patch cables. As for what preamp is the best. I cannot answer that. There are too many variables consider when designing and building your system. cheers.
Hi Mike -- I have purchased almost everything needed to make this preamp. But (here in Europe) I am not finding the 2N2907a TO-39 transistors you have specified. The TO-18 or the plastic 2n2907a are available. I have ordered some 2N2907 (no 'a'). But I just noticed that the specs are slightly different from the 'a' version. Will these work? Or can you advise a substitute that will work? Thanks -- Rob
Thanks for commenting. I’ve used both types in the past the they do work. From what I recall, I think the pin outs are different so just double check them before you solder them in place. Good luck, cheers.
Hi.. Really cool project.. Nice to see the MQ iron in use.. Have you tried a CLCRC supply? Will it better the stock supply if using a plate resistor (for starters..) ? The Edcors are nice and looks like a steal.. There are also the Cinemags, but dunno the price of those..
Thanks for commenting. I've used CLCRC in amps that don't have a CCS. It works good. You just have to be aware of its location and orientation. They can induce noise in the system. They can be expensive and if you are using a Constant Current Device you probably won't notice a difference. I've looked into Cinemags before but they seem hard to source and can be pricey. Jensen's OT's are also an option. Cheers
@@MikeFreda Thanks Mike.. BTW, I was thinking that if I first use a Plate R to try, do I also need to add a Cathode Bypass cap to get the ~same 6.4kOhm output impedance as the circuit with the CCS plate load w/ unbypassed cathode R?
You don't have to install a Cathode Resistor bypass capacitor. I never did. You can experiment if you want. 200uf to 470uf @10v. I'm not sure you would hear a difference. There are some Bypass Calculators available online. 5hz would be a good cut off point. Really no need to go lower than that.
Good eye. Those are 475k metal films that goto ground. They provide a reference to ground from the output. Known as Rg. Technically, it should effect the output impedance a little. You can experiment with it to see if you hear a difference with or without them. Cheers.
Thanks for your reply. This is a great question.I don’t prefers to be an expert on tis but I do understand the topic. When I purchased these transformers I told Mike L (at Magnequest) what I was doing and he gave me a bunch or parameters that would work well. These B7’s >200H, DCR Pri-1042ohm, Sec 37ohm. Mike told me the cap range can be between 1 - 3.3 ufd which would put it in the inaudible rage. The issue with this circuit is that the plate load is fed from a Constant Current Device which there really isn’t a formula or calculation to incorporate that. The CSS has impedance in the MOhm range. In essence we are just sizing the Parafeed cap which is a basic formula. Looking at this topic on the Tubecad site, I did rough calculations that puts it around 5hz using a 1ufd cap. It was recommended that the only “real” way to know is to try different sized caps and manually measure and graph the curve. I hope that makes sense? If you know otherwise or can add to this, please enlighten me. Cheers.
@@MikeFreda Thank You for invitation. I am not profesional so I have only suppositions Never seen in my practice such solution. Do not know it's advantages - I can deliberate only that it provides exact anode current within long time . Low power triodes last many years but why not? In usual tube amplification lower signal lowers tube current and lowers voltage drop on anode resistance (which is constant). Appears proportional higher anode voltage - phase reverse. Here tube current is stabilized in such way that total resistance of anode circuit is constant. Conclusion - the more tube resistance the less stabilizer resistance.That operates in way simmilar to push pull amplifiers except half is done by support of two semiconductors and it gains in amplification . Another speciality is no feedback which is reasonable if is sufficient linearity. Possible that the mentioned LC circuit produces at 5Hz phase shift which could make any feedback stability problematic . That is "my" theory . All best
Thanks for commenting. This is an interesting topic. C = 2xL / (RxR) is the basic formula for sizing a normal Parafeed capacitor. C is the capacitance in farads, L is the plate choke inductance in henries, and R is the load impedance seen by the plate (i.e. transformer primary impedance) in ohms. If you use kiloohms then C is in microfarads. This basic formula works well for power amps feeding speakers. The value calculated gives the best bass extension while keeping the impedance seen by the tube's plate relatively constant and resistive to the lowest possible frequency. The value is not that critical, anything from half to twice this value will work. To my understanding, there is no formula that works properly for the Constant Current Sources that are used instead of Plate Chokes. The other issue for preamps is the unloaded secondaries which complicates things even more. The simplified formula ignores the effective resistance in parallel with the OPT inductance. The approach is to experiment with different capacitor values and make measurements. You'll have to measure the frequency response at a variety of signal levels to see what works best for your speakers and ears, which I have done. I honestly cannot hear a difference between 1.0 up to 3.3. What would you recommend and why?
@@MikeFreda If I were you, I would measure frequency response and distortion. It is difficult to predict frequency dependant systems with simple theory. SPICE might help but it is just a simulation based only on user defined inputs. Totally symetrical topology might be a solution for really quiet, extremly linear and sonically neutral device. But again you will hit the issues with HQ audio grade singe ended to symetrical (and vice versa) line level transformers. It is difficult when playing on high quality level.
Sorry, this might be a very stupid question as I'm new to this. However is there a reason (except time of course) to buy a tube preamp when you can make it yourself at a much, much lower price? Perhaps it is simply too difficult to do for anyone that doesn't have extensive experience in electronics?
Thanks for commenting. This is difficult question to answer. To build this preamp as a one off build , it will probably cost you more since you had to purchase all the tools and purchase all the individual parts at retail cost. To get into tube amps as a hobby I don't think you need to have deep knowledge of electronics. If you are following some elses build or assembling a kit, basic theory is all you really need. Ohms law, safety, soldering, how to use a multi-meter, reading resistors etc... basic stuff. For scratch building, yes, you do need some knowledge. I don't build tube amps to save money. I build them because I enjoy making things. I enjoy the research, I like to learn new things, I enjoy the sounds of tubes, I like tweaking and there is this deep down thrill of making something that sounds amazing. It's the entire experience for me. Many people just don't have the tools, they are not "handy" or they are afraid they spent all the time and money and its doesn't work, which can be frustrating. I've had a few projects that just didn't work right. I chalk that up to learning. I hope that give you some insight? Cheers
Interesting single tube design. Is there a reason for using the stepped attenuator rather than just a standard variable pot? I don't feel 1% resistors are likely not needed for high quality audio. I understand you doing it for you, but even high performance tube based RF circuits didn't use 1% resistors and RF is way more fussy and susceptible to external influence than audio. Tubes are just not that fussy on accuracy, even running UL mode. My thoughts aside, this is still a great project and does look to have some great results.
Thanks for commenting. Stepped Attenuators provide better clarity, balance, longevity and are less inductive when compared to a standard variable resistor. They do cost a lot more but I do feel you do get what you pay for especially if you are critical listener especially if you use headphones. I typically use 5% resistors in my amps and only used 1% in the current setting application of the CCS. This way you get an exact value you calculated and you have a better chance to having the left and right side balanced (providing the triodes are closely matched) Cheers
1percent metal film resistor are supposed to result in less noise . how much less noise depending on the location in circuit and resistance value.i do intend to make some some comparison s.the price and increase of availability of metal film make them worth a try
I'm assuming you are asking about the Parafeed Cap sizing and its roll off? Parfeed cap size is: C = 2 x L / (R x R) C is the capacitance in farads, L is the plate choke inductance in henries, and R is the load impedance seen by the plate (i.e. transformer primary impedance) in ohms. If you calculate using kOhms the C is in microfarads. This OT has a high inductance +/-100H. My calculations put it around 5-7hz.
@@ColocasiaCorm To get the proper impedance matching. The 5670 in this circuit is set up as a voltage amplifier with an approximate output impedance of 15K. That is high to feed directly into a power amp. It would not sound good. These transformers are 15k:500ohm. You could omit the transformers but you need something like a Cathode Follower circuit to tame the higher impedance. Cheers.
How high? ~15K input impedance is pretty high for most tube amps. Perhaps Solid State amp. You would need to check the manufacturer's specs. You can also change the ratio of the output transformer to suit or add a cathode follower stage.
Thanks for commenting. Due to this circuit being a single stage (Voltage Amplifier) it would be difficult to implement that. My other preamp circuits (6SN7 or the 12AU7) that could be implemented in them. You would have to be careful with layout and topology since adding pots within the circuit could induce some noise into the system. That being said, In all my preamp designs I’ve made over the years, I’ve never thought I needed to add something like that. Cheers.
Yes/No... The PT is just a standard low current Hammond but the high capacitance in the PSU and the Constant Current Source sort of negates that fact. As for the Output Transformer, don't let the size and looks of them fool you, they are quality. Magnequest is well known in the Tube DIY circles for being one of the go to Inductor Winders. Mike purchased the rights to Peerless and all their technical documents. Wonderful stuff. Cheers
The whole point of using an output transformer from audio quality perspective is to remove the need for output coupling capacitor. You just made the sound worse by adding two of the most significant components in the signal path that will color the sound. Best is direct coupled if you can get away it. If you can’t as with most preamps, stick with either high quality output coupling capacitor (ie, no electrolytic!) or transformer, but not both. This is the first I’ve heard anyone claim an output transformer can’t block DC.
Thanks for commenting. I would agree with you that you should try to keep the components in the signal path to a minimum and/or make them as high quality as possible. This topology is called Parallel Feed and has its virtues. The main ones being the power supply characteristics are more isolated from the output. The transformer can be much smaller since it doesn’t have to handle any DC current. This allows for better high frequency response without excess interleaving and the capacitance that results. Technically, this means deeper, cleaner bass without sacrificing the highs. You do need to chose your components wisely. There is a lot of debate on this topic. I’ve built many amps and preamps the standard series feed and parallel feed and they both have their merits. I can’t really tell you which is better, they are just different. Cheers, Mike
Thanks for commenting. I find this interesting. Upon doing some research, I found that we are both right. In certain parts of the world with older english "orientated" is a common way to pronounce how things are situated from one another. I'm in Canada and its just something that a lot say without any thought. It must be the British influence. lol Cheers.
Thanks for commenting, Unfortunately I don't really sell my projects/gear. Part of the issue is that I have to pay full retail for all the components and at that cost you may be able to buy something commercially available or a proper tube amp "kit". Cheers.
Thanks for the discussion on stepped attenuators, and the reference to GoldPoint.
Thanks for commenting. I've been happy with my Goldpoint attenuators and would purchase from them again. Their website also has some good resource material for DYI'ers. Cheers.
Been using a 5670 tube buffer PCB kit now for a year or two, and I absolutely love the thing! I wanna do a similar project, just without the bread boards. I'm gonna grab an input relay/remote volume kit, soft on/off board and a PSU in a similar enclosure. Super nice project here, great job!
Thanks for the comment. Cheers.
Thanks for commenting. This tube is a really sleeper. Inexpensive and obtainable. Cheers.
@@MikeFreda I discovered a message board post going back where someone had done all the leg work, testing almost every tube of this type out there, and concluded the JAN GE milspec tubes had the most pleasing sound overall, mainly in the bass dept. I had my reservations with the dinky kit I got as well, but low and behold I prefer it in my system than without, which ultimately lead me to researching how I can make it into a full legit linestage and not some PCB sandwitched between pieces of acrylic.
I heard the golden words - dead quiet - I'm interested!
Yes, my pet peeve… a “noisy” amp.
My ears perked up as well. I can deal with hiss on old recordings but I want my gear quiet, quiet. This looks like a really interesting build.
@@bugdrvr try make kondo m77 clone with 12AY7.less gain than ecc83.pretty good phono stage..but be true-need good parts.china made really poor..
CCS has always mystified me. You made it very clear. Thank you!
Thanks for commenting. You're welcome. I have some more information on CCS on my 6DN7 amp build (Part 3 discusses CCS) Cheers.
This badboy tube is really great in SRPP circuit
Would definitely love to make one of these.
Thanks for commenting. If you do, I’d like to hear how you make out.
Very nicely built and explained. Keep it up Mike. Regards from 🇮🇳
Thank you. Cheers.
This is great, thanks. I built all solid state gear. Wanted a tube pre but the gain is usually to high and output impedance didnt match amp. Might give this a go.
Thanks. Hope you make it and it turn out! Cheers.
5670 is a great tube. I have a design for 5670 preamp. Works great. The jan tubes are great
Yes, it is. One of my favourite and they are surprisingly affordable. What sort of topology do you use?
Wow this looks really fun to build.
Thank you! Cheers.
Happy New year 2023 👍 thanks for your new VT Pre Amp ......hope I will try .....🌹💗
Thank you for commenting. Happy New Year to you as well. Cheers.
thank you for the video! I'm interested in trying this build out. Have never built an amplifier before so this should be a lot of fun.
Glad you like the video. This pre-amp build is a bit more complicated and costly than my 12AU7 or 6SN7 preamp. Have a look at those videos.
Happy New year! Thanks for another in-depth build!
You're welcome, Happy New Years to you as well! Cheers!
This might be a bit elementary (an am a bit of a novice) but it may help me and other to see a video of you assembling the perf board
I’m assuming you are inquiring about the CCS board. I have another video about installing CCS on my 6DN7 amp. I go into detail about this board, how it works and also have a downloadable worksheet. Cheers.
@@MikeFreda OH. OK Mike. Thank you very much
Just so happen I have two Trigon brand 6SN7GT tubes in my stash of tubes, was thinking of using them to build an OTL headphone amp at one stage.
Interesting, you will have to get a bit creative to adapt 6SN7 to a OTL headphone amp. You may need another tube. Love to hear how you make out. Cheers.
Perfect sharing thanks you for detail. If tube change ecc83 can use?
Not really. Do not recommend. A 12AX7 has too much gain (mu of 100) for this application and the output impedance is also different.
You would better luck with a 12AU7 or a 12AY7 but you would need to select different output transformers due to the higher output impedance of these tubes.
Sort answer, no. It’s the wrong tube for the job. Probably be noisy as well. Cheers.
Can I make a suggestion?
As a recording engineer and musician, who wants to build one of these myself
May I suggest that you have dual functionality one for recording and one for monitoring through your daw digital audio workstation
As I believe the tubes need to be somewhere else in the circuit to do these two jobs separately, which can be easily fixed as long as the box is set up right from the beginning
So then you want to monitor your mix when you’re not recording through the tubes
Generally, for mastering purposes
Interesting concept. I'd have to research and study that. Does something like this already exist in a tube format?
@@MikeFreda I don’t think so
I am definitely not a tube expert myself. I have some audio engineering skills, but of the electrical nature none
Although I am very interested in the subject
But I can point to do a video where somebody mentioned that concept that I am talking about
@@MikeFreda ua-cam.com/video/j2vPU4ShL-c/v-deo.htmlsi=CilvauSVxW2FrBoe
This is the guy I was listening to
Wow you are so talented. I wish I could do that.
Thanks for commenting. If you research, readup and start small, you could do something like this. Cheers
This is absolutely fabulous ♦️
nice do have some mic tube preamp that can distortion? non-transistor project
What are you trying to accomplish? What would be the application? Studio input from a guitar?
Excellent...
Not to easy to find that high quality part in my local market (even in my country). Just one thing, why that headphone jack sets in the back side?
Yes, this preamp design does require quality components to make it sound good. The headphone Jack is an afterthought. I wasn’t sure if it would work. It was just easy to install it right beside the output…. But yes, it would easier to use if it was on the front. I would use shielded cables. That being said, I don’t really use this as a headphone amp (I already have a dedicated one) so it’s just there for proof of concept and to see if it could drive high impedance headphones, and it can. Cheers.
Very nice build, as always. Do you mind sharing sourcing for internal wire including your shielded cable, please? Thank you!
Hello, thanks. I'm fortunate to have a really good electronic store near me and I'm able to get a lot of my supplies from. This is some of the hook up wire I use: leeselectronic.com/qc/product/224902-hook-up-wire-22awg-solid-ul1007-red-50ft-spool.html
They also sell 2 wire, shielded wire from spools. leeselectronic.com/qc/product/23814-high-voltage-2x22awg-w-ground-ft4-jacked.html
I've found similar wire at Mouser but they typically have spools and it can get pricey, cheers.
Exzellent.
Thanks. Glad you like the build. Cheers.
youre getting damn good at this mike! love it ❤
Thank you! Hope you have a Happy New Year! Cheers
Very nice thank you for sharing. Why did you choose the 100KOhm volume control over 10KOOhm version?
The input impedance of the 5670 is around 100K. I mostly use a turntable (with a phono stage) and a NAD CD player which they typically like to see 50K or higher. Their output levels seem to pair well with it this pre-amp design. You could try 10K if you like. I'm not sure what the end effect will be. Cheers
Well done, finally line stage with appropriately low gain, unlike all those ECC83 stages having way too much gain and weak drive (good for 100k but not 600r). I like use of veroboards because they allow for optimal layout with either small pins or tagboards.
Tube is biased very hot. Didn't do calculations for 600r load like headphones but they might do fine with lower dissipation when 15k:600 OT's are used, i guess such biasing was used for the sound.
I'm very interested in seeing turret or vero board layout for two stages of class a push pull circuit with tubes mounted horizontally. If anyone ever made mic or phono preamp of this type please let me know, layout for vertically mounted tubes would also help because it is not that different in terms of layout.
Thanks for commenting. Max dissipation for a 5670 is 16.5ma. I'm running around 8ma per section which is 1/2 of max and aligns with the 150v curve. As an input tube you would go way less than that. It's a balance of gain and tube longevity. Cheers!
@@MikeFreda Yes, i should look at suggested biasing instead of max rating. Btw, 5670 was often used as vari gm tube for compressors with mixed results because it was not really made for this application like 6386 and other were. 6CG7 is another interesting tube for line stages very similar to 6SN7 in 9 pin bottle, others like 6N6, 12BH7 and similar small signal drivers would also do fine where gain should be very low.
Is CCS really needed in class a where current is pretty much constant? Anyway, maybe try really high quality transformers like Jensen with 80% Ni cores, Lundahl has their own special cores too, although low mid distortion is different from Fe/Ni cores.
@@Signal_Glow No, you don't need the CCS, you could replace it with a plate choke. I would suggest something with the appropriate Ma rating and a large inductance (i.e.40H or more). The larger, the better... but they will get costly. You will have to know what the DCR rating is and recalculate the B+ voltage to give you ~150V at the plate. You could also change the topology from Parallel Feed to a Series Feed and use normal OT but you will sacrifice some performance and you may lower the noise floor a bit as well. The CCS does a great job of PSRR. At the end of the day, the CCS will be your most cost effective option and will give you the best performance. Cheers
Nice work. I'm wondering if this would sound better than my Wavelength Sine Preamp? It's an early version from 1994-95 using 6SN7's, Jensen input transformers, Penny&Giles volume pot and non-regulated power supply with 6X5 rectifier. There are two Hovland coupling caps in each channel. I've contemplated replacing the Hovland caps with Mundorf Silver/Oil because swapping them out in the Cardinal 300B SET amps made quite a difference. Was also wondering if a stepped attenuator would be an improvement. Maybe it would be better to build something from scratch?
Reply
Thanks for commenting. It would be very hard to say which is "better" . I think they would sound different but better, not sure. Upgrading caps and a better volume control can add come clarity. 6SN7's have a reputation for sounding "tubey" with a strong midrange presence (at least that's what people feel) They were the go to preamp tube in the 80's and 90's and then just like that, people started to shy away from them. I have a 6SN7 pre as well but the topology and gain is different so it's hard to tell. This 5670 pre is much quieter. That goes a long way for me to sounding better. I like tube rectifiers in guitar amps but I feel solid state rectification in HiFi is preferred since you can easily incorporate higher value caps in the PSU, creates less heat, last longer, little to no voltage drop across the diode and its cost less. (some people also feel there is no sag either but I have no evidence in that since you would need to build similar circuits and test that theory) Cheers, Mike
@@MikeFreda Thanks for your reply. I thought Golden Dragon 6SN7s sounded great in my preamp. At the time when Gordon Rankin built the preamp, he told me he thought the Penny & Giles Pot sounded as good as any stepped attenuator he'd heard. And it was quite expensive. The caps are another story since there are many more choices available now. Several years ago, per Gordon's recommendation, I swapped out the Hovlands in my Cardinal 300B amps. The Mundorf Silver/Oil sound incredible. Though, Mundorfs for the preamp would be over $300. So I may audition other preamp topologies and tube types before upgrading. I know Gordon had moved on from 6SN7s. Probably be easier to build something from scratch than making major changes to this preamp. Thanks again.
Great preamp. Do you have schematic for use power transformer without transistor current source. Thanks
Unfortunately, I don't. The performance of this circuit design would not be great if you removed the Constant Current Source. You could try and remove the CCS and install a ~14,000 ohm 2 watt Plate Resistor and use a 240 ohm Cathode Resistor but I'm not sure if there enough voltage swing to make that work properly. Worth a try I guess. If you want to use this tube without a CCS you basically need to rework the PSU to have higher B+ (Higher voltage PT) and better channel separation buy using an extra RC node for each channel or use some chokes. I'd recommend looking at the 5670 Tube Data Sheets for more information. I'm not sure if there is much of an appetite for me to rework this circuit without the CSS. It would be easier to build but it would be a step backwards, cheers
@@MikeFreda thank you so much
@@MikeFreda so you are saying I will try 14kohm resistor with 150 volts supply?
No, that's not what I am saying. Keep the PSU the same and just replace the CCS with a 14k 2W resistor. The tube needs 150V at the plate to operate properly. To get enough voltage swing, you need the B+ to be at least 100V more. Last cap at the PSU ~265VDC - 14K resistor 150VDC at the plate of the 5670. To be clear, I have not done this. I'm not sure how it will sound. Cheers.
@@MikeFreda i understand, I will try that. Thank you so much
Why didn't you make the output balance? since you already use transformer...
Thanks for your comment. Yes, you can with these transformers and circuit. I have done this in the past but I didn’t really hear a difference into my 2A3 power amps, speaker and room setup. I suspect one would definitely hear a difference into headphones. Cheers.
Can this unit be built for a recording studio scenario?
For mic’s and guitar input? No, not enough gain and the input impedance are not ideal. For the output impedance, you need to see what gear you are chaining to. Could this circuit be modified, sure but you need more information. Cheers.
What do you think of Mr. Carlsons Lab Mike? Do you follow his videos, I learn a lot from you and him both. You seem like you learned electronics, Carlson seems like he eats, lives, and breathes electronics. You and Carlson are what I aspire to be. Thanks for the great vids my man.
Thanks for commenting. Yes, I know Mr Carlson and enjoy his in-depth videos as well. His channel is his life (along with his Patreon section) and he’s really focused on electronics. Great stuff. My channel is mostly about making and it’s just a hobby for me. I enjoy sharing my projects and knowledge to the world. Please let me know if there is other content you would like to see. Cheers.
Now... how much more complex would this get should one want also a tape loop in this thing?
This would be complex to incorporate in this preamp. There isn’t enough gain and you would also require an equalization network. What I would recommend is a separate tape preamp before this preamp. You could make a Phono style preamp and modify the RIAA LP equalization network with NAB 1/4” 7.5/15ips equalization network. That mod would be easy enough to do, just need to change some capacitor and resistor values. Cheers.
Must ask, where do you get the selector knobs?
Thanks for commenting. I made the knobs from scratch. I have a metal lathe. Please have a look at my shop tour video and my other tube amp videos to see how I fabricate most of this items. Cheers.
Nice…..can you think about doing a preamp that also has an option for tone controls? In my room I always find a bit of bass lift is useful.
You could probably add tone controls to this one quite easily using passive tone control circuitry, though you may find you don't need that depending on what your source material is. Every amp has its own signature and personality. Also will depend how you match it to your power amp. You could put an equalizer between this amp and your power amp if you need to compensate for room errors.
You can but with a single tube design it is a bit harder and you will probably lose some gain. A different topology would lend itself better to tone control. Cheers
Sure !
I have a great integrated but if I were to try a pre I would think a great preamp is the way to go. What is the best? I am in Europe and have no workshop to do the handwork needed. I need to learn more.
Thanks for commenting. I believe having separate components is slightly better mostly due to the power supplies being separate. One issue to overcome is injecting noise into the system due to ground loops and patch cables. As for what preamp is the best. I cannot answer that. There are too many variables consider when designing and building your system. cheers.
Hi Mike -- I have purchased almost everything needed to make this preamp. But (here in Europe) I am not finding the 2N2907a TO-39 transistors you have specified. The TO-18 or the plastic 2n2907a are available. I have ordered some 2N2907 (no 'a'). But I just noticed that the specs are slightly different from the 'a' version. Will these work? Or can you advise a substitute that will work? Thanks -- Rob
Thanks for commenting. I’ve used both types in the past the they do work. From what I recall, I think the pin outs are different so just double check them before you solder them in place. Good luck, cheers.
Hi.. Really cool project.. Nice to see the MQ iron in use.. Have you tried a CLCRC supply? Will it better the stock supply if using a plate resistor (for starters..) ? The Edcors are nice and looks like a steal.. There are also the Cinemags, but dunno the price of those..
Thanks for commenting. I've used CLCRC in amps that don't have a CCS. It works good. You just have to be aware of its location and orientation. They can induce noise in the system. They can be expensive and if you are using a Constant Current Device you probably won't notice a difference. I've looked into Cinemags before but they seem hard to source and can be pricey. Jensen's OT's are also an option. Cheers
@@MikeFreda
Thanks Mike.. BTW, I was thinking that if I first use a Plate R to try, do I also need to add a Cathode Bypass cap to get the ~same 6.4kOhm output impedance as the circuit with the CCS plate load w/ unbypassed cathode R?
You don't have to install a Cathode Resistor bypass capacitor. I never did. You can experiment if you want. 200uf to 470uf @10v. I'm not sure you would hear a difference. There are some Bypass Calculators available online. 5hz would be a good cut off point. Really no need to go lower than that.
great thanks.
You're welcome! Cheers.
what are the two other resisters on the outputs i see the two 10k for the subs it looks like the other two go to the ground buss
Good eye. Those are 475k metal films that goto ground. They provide a reference to ground from the output. Known as Rg. Technically, it should effect the output impedance a little. You can experiment with it to see if you hear a difference with or without them. Cheers.
Nice impression. What is the the serial resonance of output tr primary winding which is not stated and 1uF capacitance? Is it beyond acoustic range ?
Thanks for your reply. This is a great question.I don’t prefers to be an expert on tis but I do understand the topic. When I purchased these transformers I told Mike L (at Magnequest) what I was doing and he gave me a bunch or parameters that would work well. These B7’s >200H, DCR Pri-1042ohm, Sec 37ohm. Mike told me the cap range can be between 1 - 3.3 ufd which would put it in the inaudible rage. The issue with this circuit is that the plate load is fed from a Constant Current Device which there really isn’t a formula or calculation to incorporate that. The CSS has impedance in the MOhm range. In essence we are just sizing the Parafeed cap which is a basic formula. Looking at this topic on the Tubecad site, I did rough calculations that puts it around 5hz using a 1ufd cap. It was recommended that the only “real” way to know is to try different sized caps and manually measure and graph the curve. I hope that makes sense? If you know otherwise or can add to this, please enlighten me. Cheers.
@@MikeFreda Thank You for invitation. I am not profesional so I have only suppositions Never seen in my practice such solution. Do not know it's advantages - I can deliberate only that it provides exact anode current within long time . Low power triodes last many years but why not?
In usual tube amplification lower signal lowers tube current and lowers voltage drop on anode resistance (which is constant). Appears proportional higher anode voltage - phase reverse.
Here tube current is stabilized in such way that total resistance of anode circuit is constant. Conclusion - the more tube resistance the less stabilizer resistance.That operates in way simmilar to push pull amplifiers except half is done by support of two semiconductors and it gains in amplification .
Another speciality is no feedback which is reasonable if is sufficient linearity. Possible that the mentioned LC circuit produces at 5Hz phase shift which could make any feedback stability problematic . That is "my" theory . All best
Capacitors on output matching transformers are too small for parafeed topology.
Thanks for commenting. This is an interesting topic. C = 2xL / (RxR) is the basic formula for sizing a normal Parafeed capacitor. C is the capacitance in farads, L is the plate choke inductance in henries, and R is the load impedance seen by the plate (i.e. transformer primary impedance) in ohms. If you use kiloohms then C is in microfarads.
This basic formula works well for power amps feeding speakers. The value calculated gives the best bass extension while keeping the impedance seen by the tube's plate relatively constant and resistive to the lowest possible frequency. The value is not that critical, anything from half to twice this value will work.
To my understanding, there is no formula that works properly for the Constant Current Sources that are used instead of Plate Chokes. The other issue for preamps is the unloaded secondaries which complicates things even more. The simplified formula ignores the effective resistance in parallel with the OPT inductance. The approach is to experiment with different capacitor values and make measurements. You'll have to measure the frequency response at a variety of signal levels to see what works best for your speakers and ears, which I have done. I honestly cannot hear a difference between 1.0 up to 3.3.
What would you recommend and why?
@@MikeFreda If I were you, I would measure frequency response and distortion. It is difficult to predict frequency dependant systems with simple theory. SPICE might help but it is just a simulation based only on user defined inputs. Totally symetrical topology might be a solution for really quiet, extremly linear and sonically neutral device. But again you will hit the issues with HQ audio grade singe ended to symetrical (and vice versa) line level transformers. It is difficult when playing on high quality level.
Sorry, this might be a very stupid question as I'm new to this. However is there a reason (except time of course) to buy a tube preamp when you can make it yourself at a much, much lower price? Perhaps it is simply too difficult to do for anyone that doesn't have extensive experience in electronics?
Thanks for commenting. This is difficult question to answer. To build this preamp as a one off build , it will probably cost you more since you had to purchase all the tools and purchase all the individual parts at retail cost. To get into tube amps as a hobby I don't think you need to have deep knowledge of electronics. If you are following some elses build or assembling a kit, basic theory is all you really need. Ohms law, safety, soldering, how to use a multi-meter, reading resistors etc... basic stuff. For scratch building, yes, you do need some knowledge. I don't build tube amps to save money. I build them because I enjoy making things. I enjoy the research, I like to learn new things, I enjoy the sounds of tubes, I like tweaking and there is this deep down thrill of making something that sounds amazing. It's the entire experience for me. Many people just don't have the tools, they are not "handy" or they are afraid they spent all the time and money and its doesn't work, which can be frustrating. I've had a few projects that just didn't work right. I chalk that up to learning. I hope that give you some insight? Cheers
@@MikeFreda Thanks this does help and it's cool to know you're just doing it for the thrill of it.
❤❤❤❤❤
Interesting single tube design. Is there a reason for using the stepped attenuator rather than just a standard variable pot? I don't feel 1% resistors are likely not needed for high quality audio. I understand you doing it for you, but even high performance tube based RF circuits didn't use 1% resistors and RF is way more fussy and susceptible to external influence than audio. Tubes are just not that fussy on accuracy, even running UL mode. My thoughts aside, this is still a great project and does look to have some great results.
Thanks for commenting. Stepped Attenuators provide better clarity, balance, longevity and are less inductive when compared to a standard variable resistor. They do cost a lot more but I do feel you do get what you pay for especially if you are critical listener especially if you use headphones. I typically use 5% resistors in my amps and only used 1% in the current setting application of the CCS. This way you get an exact value you calculated and you have a better chance to having the left and right side balanced (providing the triodes are closely matched) Cheers
Has better clarity at lower levels especially with a shunt attenuator (as opposed to series..)
1percent metal film resistor are supposed to result in less noise . how much less noise depending on the location in circuit and resistance value.i do intend to make some some comparison s.the price and increase of availability of metal film make them worth a try
Do you have any idea which may be the resonance frequency of the output group, capacitor+OT ?
I'm assuming you are asking about the Parafeed Cap sizing and its roll off? Parfeed cap size is: C = 2 x L / (R x R)
C is the capacitance in farads, L is the plate choke inductance in henries, and R is the load impedance seen by the plate (i.e. transformer primary impedance) in ohms. If you calculate using kOhms the C is in microfarads. This OT has a high inductance +/-100H. My calculations put it around 5-7hz.
@@MikeFreda Thank you! If that's so there's nothing to worry about the linearity of the output stage.
Would love a layout if you have the time
Link for the schematic and layout is in the description.
❤❤🎉
Thanks
Why do you use an output trafo for line level
Very interesting. Step down.
@@ColocasiaCorm To get the proper impedance matching. The 5670 in this circuit is set up as a voltage amplifier with an approximate output impedance of 15K. That is high to feed directly into a power amp. It would not sound good. These transformers are 15k:500ohm. You could omit the transformers but you need something like a Cathode Follower circuit to tame the higher impedance. Cheers.
@@MikeFreda what if the input impedance you were feeding to was higher?
How high? ~15K input impedance is pretty high for most tube amps. Perhaps Solid State amp. You would need to check the manufacturer's specs. You can also change the ratio of the output transformer to suit or add a cathode follower stage.
hi
is it possible to add bass and travel gain to this design? can I contact you via mail?
Thanks for commenting. Due to this circuit being a single stage (Voltage Amplifier) it would be difficult to implement that. My other preamp circuits (6SN7 or the 12AU7) that could be implemented in them. You would have to be careful with layout and topology since adding pots within the circuit could induce some noise into the system. That being said, In all my preamp designs I’ve made over the years, I’ve never thought I needed to add something like that. Cheers.
It's so hard to find chassis in Europe, even from China the cost is 100€ or more.
That’s too bad. You could adapt this circuit to an aluminum top plate with a wooden base. I’ve made many amps this way. Cheers.
Hi. Is this point to point
Yes/No... half the amp is point to point the other parts are mounted on PCB's. You could mount the PSU on standoffs but not the CCS. Cheers
Hi Mike, You must be in. Vancouver too……Would love to chat!
Yes… near Vancouver, Langley. Sure, please PM me if you like. Thx.
Those transformers don't look very high fidelity
Yes/No... The PT is just a standard low current Hammond but the high capacitance in the PSU and the Constant Current Source sort of negates that fact. As for the Output Transformer, don't let the size and looks of them fool you, they are quality. Magnequest is well known in the Tube DIY circles for being one of the go to Inductor Winders. Mike purchased the rights to Peerless and all their technical documents. Wonderful stuff. Cheers
It is “attenuator” not “antenuator”
Right, good catch! Thanks, Mike.
The whole point of using an output transformer from audio quality perspective is to remove the need for output coupling capacitor. You just made the sound worse by adding two of the most significant components in the signal path that will color the sound. Best is direct coupled if you can get away it. If you can’t as with most preamps, stick with either high quality output coupling capacitor (ie, no electrolytic!) or transformer, but not both. This is the first I’ve heard anyone claim an output transformer can’t block DC.
Thanks for commenting. I would agree with you that you should try to keep the components in the signal path to a minimum and/or make them as high quality as possible. This topology is called Parallel Feed and has its virtues. The main ones being the power supply characteristics are more isolated from the output. The transformer can be much smaller since it doesn’t have to handle any DC current. This allows for better high frequency response without excess interleaving and the capacitance that results. Technically, this means deeper, cleaner bass without sacrificing the highs. You do need to chose your components wisely. There is a lot of debate on this topic. I’ve built many amps and preamps the standard series feed and parallel feed and they both have their merits. I can’t really tell you which is better, they are just different. Cheers, Mike
The word is "oriented", not "orientated".
Thanks for commenting. I find this interesting. Upon doing some research, I found that we are both right. In certain parts of the world with older english "orientated" is a common way to pronounce how things are situated from one another. I'm in Canada and its just something that a lot say without any thought. It must be the British influence. lol Cheers.
My mother said, "If you can't say something nice, don't say anything at all."
videos, plural. No apostrophe. ideas, plural. No apostrophe.
Thanks for bringing that to my attention. Cheers.
Hey Mike, do you want to sell one ????? I'm also in Canada.
Thanks for commenting, Unfortunately I don't really sell my projects/gear. Part of the issue is that I have to pay full retail for all the components and at that cost you may be able to buy something commercially available or a proper tube amp "kit". Cheers.