Pars Posted September 18, 2020 Report Posted September 18, 2020 This topic has needed its own thread for quite some time (years), so starting one. I may see if I can get a mode to move some of the posts here (Carbon thread, not sure where else they are littered). 11 hours ago, Pars said: GRHV question: I'm building a pair of the singles intended for 400V output. On the BOM I have, there are 2 680uf caps for each board; 1 is a 550V caP, the other a 450V cap. I presume the 550V would go right after the rectifiers in the schematic, and the 450V on the output. The BOM that was posted by jamesmking uses 4 x 470uf 550V Kemets. Is there much difference between the 680uf and 470uf in terms of sound or anything? BTW, I think I had mentioned this before, but the GRHV really deserves its own thread. Confusing to have to weed through the Carbon thread, and I don't know how many other threads with GRHV posts in them. 2
starcat Posted February 20, 2021 Report Posted February 20, 2021 So, what's up with this dedicated GRHV thread? No one building yet a GRHV 🙂
jamesmking Posted February 20, 2021 Report Posted February 20, 2021 8 hours ago, starcat said: So, what's up with this dedicated GRHV thread? No one building yet a GRHV 🙂 1. part of the headcase experience is trawling a massive number of posts in one thread e.g. the carbon to find out about the grhv. Having a dedicated thread is seen as cheating. 🙂 2. to build a grhv you need C2M1000170D which are out of stock just about everywhere with little hope of being back in stock anytime soon... this is one of the issues delaying my megatron build. On 9/18/2020 at 3:10 PM, Pars said: This topic has needed its own thread for quite some time (years), so starting one. I may see if I can get a mode to move some of the posts here (Carbon thread, not sure where else they are littered). GRHV question: I'm building a pair of the singles intended for 400V output. On the BOM I have, there are 2 680uf caps for each board; 1 is a 550V caP, the other a 450V cap. I presume the 550V would go right after the rectifiers in the schematic, and the 450V on the output. The BOM that was posted by jamesmking uses 4 x 470uf 550V Kemets. Is there much difference between the 680uf and 470uf in terms of sound or anything? BTW, I think I had mentioned this before, but the GRHV really deserves its own thread. Confusing to have to weed through the Carbon thread, and I don't know how many other threads with GRHV posts in them. I can't comment on any sound quality differences, but there very few options if you want a 680uF cap that is 550V and will fit in a 2u case, i.e. be about 65mm in height or less. 450V or better 500V would be fine for the output cap with 400V output. The + lead of the input cap is connected to the centre pin of the large C2M1000170D. Output cap + lead is connected to the cathode of the IN4007 protection diode that is directly inline with the 5.1ohm 3W current sense resistor. 1
JoaMat Posted February 20, 2021 Report Posted February 20, 2021 (edited) 1 hour ago, jamesmking said: 2. to build a grhv you need C2M1000170D which are out of stock just about everywhere with little hope of being back in stock anytime soon... this is one of the issues delaying my megatron build. Most, if not all, public published GRHV boards are with c2m1000170d but if a bit creative you might use fqpf8n80c. If you do - please tell us how it turned out. We have to find ways to tackle unavailable parts. Edited February 20, 2021 by JoaMat 1
Kerry Posted February 20, 2021 Report Posted February 20, 2021 I’ve been using fqpf8n80c with my miniature miniaturized (smaller than the group buy version) Golden reference supplies. I also have the negative supply GRHV79 working in there too. I always use the external current regulators now since they are much faster and actually protect the supply. Works great. I’ve got a lot of hours on these. 6
starcat Posted February 20, 2021 Report Posted February 20, 2021 13 hours ago, jamesmking said: 2. to build a grhv you need C2M1000170D which are out of stock just about everywhere with little hope of being back in stock anytime soon... this is one of the issues delaying my megatron build. The C2M1000170D will be available soon here https://at.rs-online.com/web/p/mosfet/1456660//?cm_mmc=aff2-_-at-_-octopart-_-Outofstock1456660 and here https://www.newark.com/wolfspeed/c2m1000170d/power-mosfet-n-channel-4-9-a-1/dp/12X8366?CMP=AFC-OP
iemphile Posted February 21, 2021 Report Posted February 21, 2021 (edited) I ordered some C2M1000170D from Digikey in January, estimated ship date was 3/19/2021. Edited February 21, 2021 by iemphile
jamesmking Posted February 21, 2021 Report Posted February 21, 2021 (edited) My guide to golden reference high voltage (grhv) variants Here is my attempt at documenting some of the different GRHV board variations. (this will be work in progress for sometime) feel free to private message me with any additions comments, corrections etc. please note the schematic for version 1.5 has been reverse engineered by me from the gerbers and may contain error(s). location of the gerbers: https://drive.google.com/drive/folders/0B_iJFfZStuVhSE5nOHBVdTByR1k golden reference filename naming guide fat - boards which are wider and have less height than the non fat variants L bracket mounting only same as the non fat variants. fat with S as the last character of the filename - boards which are wider and have less height than the non fat variants and the height is about 6.3mm less than the fat variant. these boards also feature more screw hole options for the mounting the pcb to the case including an option of not using an L shaped bracket and mounting the transistors directly to the vertical face of a heatsink/side of the case. a silkscreen cut line is included to guide removal of the L bracket mounting roles to facilitate direct transistor to heatsink mounting. dual - has both positive and negative hv supply with additional bias and both positive and negative low voltage supplies on a single pcb left - has a single negative hv rail and both positive and negative low voltage supplies right - has a single positive hv rail and bias no new in the filename version - 1.5 boards high voltage starts instantly board has high voltage VAC power. Topology is much simpler then version 1.7 and 1.8 and therefore not considered golden reference. new in the filename actual golden reference boards with extensive post pass transistor regulation. new without sw in the filename- version 1.7 high voltage starts instantly board has high voltage VAC power. new with sw in the filename - version 1.8 or later boards same topology as 1.7 but high voltage will not start until dc power supplied to the cpc1117n no dual and no fat in the filename - minimal width and quite high. NOTE placement of the screw terminals for input VAC can vary considerably depending on if the board is a fat, fat S or non fat. Overview of the versions It is the extensive post pass transistor regulation that marks the golden reference out from the other power supply designs as such version 1.5 and earlier not classify as a golden reference since they lack any post pass transistor regulation. No details of versions before 1.5 or version 1.6 could be found in the gerber archives. Apparently pre 1.5 versions had on board heatsinks and ground lanes but there gerbers and not available in the google drive archive. No schematic could be found for version 1.5 although after reverse engineering it was found that the circlotronps.pdf is similar in topology but modified to support 900V output. No bill of materials could be found for version 1.5 and it is not certain what is the power rating for the larger resistors. All versions have the pcb tracks on the underside with no tracks on the top. All versions have no ground plane ALL versions have bleed resistors to help discharge the high voltage caps although version 1.5 and a few variants of 1.7 use a single high power resistor and the rest two 1/2W resistors in series. All versions have the same basic voltage regulator based low voltage positive and negative supplies All versions have the same bias supply although v1.5 has different resistor values. ALL versions use a LT1021-10 voltage reference All the versions 1.5, 1.7 and 1.8 use a C2M1000170D pass transistor All the versions use a 10M90S pre regulator although version 1.5 controls it with a zener string and 1.7 and 1.8 control with a DN2540. 1.7 and 1.8 are almost identical in topology although 1.8 does not automatically start the high voltages. DC must be applied to the CPC1117N to start. note 1.8 does not have provision for screw terminals for the DC to switch on the cpc1117n. versions 1.7 and 1.8 have more post regulation using a second 10m90s (which is not present in version 1.5) after the 1N4007 protection diode. version 1.5 uses a 10ohm current sense resistor and versions 1.7 and 1.8 use 5.1ohm by default. version 1.5 uses a 10K resistor to ground in the 3 series resistor chain to decide the output voltage, 1.7 and 1.8 use 20K to get 400V output. The other 2 resistors in the chain are the same for version 1.5 and version 1.7/1.8 version 1.5 uses two stn0214 which are replaced by two KSC5026M in v1.7/1.8 version 1.5 uses two stn9360 which are replaced by two BC557B in v1.7/1.8 Known non golden reference boards which are variations or predecessors to version 1.5 and golden reference variants NOTE These have not been looked at in detail by me or fully reverse engineered. The cyclotron power supply series follows a similar evolution to the golden reference. earlier cyclotron psus (circlotronps.PDF) use a variation of version 1.5 topology with modifications to bring the output up to 900V. However since it lacks any post pass transistor regulation it can not be considered a golden reference these psus do not have new in their name. The later cyclotron power supplies have the words new or newver in their name on quick inspection of the gerbers appear to be modifications for 900V of the version 1.7 power supplies, have post pass transistor regulation and may be considered golden reference variants. The filename conventions follows the guide above with dual, single, S and SWS versions available. circlotronhvpowerdualnew.zip appears to be derived from the version 1.7 topology with a similar complement of transistors. On the underside of the pcb there are two additional stn0214 transistors compared to the version 1.7 plus 3 mount points for a total of 500V worth of smd zeners. circlotronhvpowerdualnewer2.zip ditches the zener string and introduces another stn3960 and includes sw and sws variations. The kgsshv8g.pdf power supply appears to be an predecessor version of version 1.5. There are very close similarities in topology including no post pass transistor regulation. Current sense is still controlled by a 2n3904. The pre-regulator is 10m90s controlled by a zener string and a quad of transistors like the version 1.5. However the pass transistor is two parallel fqpf8n80c. The control transistors are 2sa1486 instead of stn9360 and 2sc3840 instead of stn0214. Both types of control transistor are no longer in production and hence the conclusion this is older than version 1.5. Apparently these boards are classified as KGSSHV ------------------------------------------------ Version 1.5 boards ------------------------------------------------- predecessor to the golden reference with no post pass transistor regulation the board uses a single high wattage bleed resistor for each cap uses multiple stn9360s, stn0214s uses a serial zener string for to control the 10m90s pre regulator. LT1021-10 voltage reference based single sided board kgsshvpssicfetdual2.zip this has both positive and negative rails, a simple voltage regulator based low voltage positive and negative supply and 580V for stax headphones bias board size approx. width 152.7mm height 165mm silkscreen revision 1.5 kgsshvpssicfetsingle2.zip this has a single rail identical to the kgsshvpssicfetdual2 in topology and 580V for stax headphones bias board size approx. width 76.5mm height 165mm silkscreen revision 1.5 ------------------------------------------------ Version 1.7 boards ------------------------------------------------- considered the first known golden reference board version (given the lack of information on version 1.6) has extensive post pass transistor regulation. the board uses two serial bleed resistors for each cap uses a single stn9360s and multiple BC557B and KSC5026M uses dn2540 and 10m90s for stage early voltage control for the input to the c2m1000170D LT1021-10 voltage reference based double sided board kgsshvpssicfetdual2new.zip this has both positive and negative rails, a simple voltage regulator based low voltage positive and negative supply and 580V for stax headphones bias board size approx. width 152.7mm height 165mm silkscreen revision 1.7 kgsshvpssicfetsingle2new.zip this has a single positive rail identical to the kgsshvpssicfetdual2new in topology and 580V for stax headphones bias board size approx. width 76.5mm height 165mm silkscreen revision 1.7 single bleed resistor for each cap kgsshvpssicfetsinglenewleft.zip this has a single negative rail identical to the kgsshvpssicfetdual2new in topology and has a simple low voltage positive and negative regulators based supply board size approx. width 76.4mm height 165.3mm silkscreen revision 1.7 kgsshvpssicfetsinglenewright.zip this has a single positive rail identical to the kgsshvpssicfetdual2new in topology and bias supply board size approx. width 76.4mm height 165.3mm silkscreen revision 1.7 kgsshvpssicfetsinglenewleftfat.zip this has a single negative rail identical to the kgsshvpssicfetdual2new in topology and has a simple low voltage positive and negative regulators based supply board size approx. width 138.7mm height 103.1mm silkscreen revision 1.71 kgsshvpssicfetsinglenewrightfat.zip this has a single positive rail identical to the kgsshvpssicfetdual2new in topology and has a 580v stax bias supply board size approx. width 138.6mm height 103.1mm silkscreen revision 1.71 kgsshvpssicfetsinglenewleftfatS.zip this has a single negative rail identical to the kgsshvpssicfetdual2new in topology and has a simple low voltage positive and negative regulators based supply has more mounting options for the pass transistor, 4 holes for mounting to a case and can be used with or without an L shaped bracket board size approx. width 138.7mm height 98.6mm (slightly less height than the fat non S version silkscreen revision 1.71s kgsshvpssicfetsinglenewrightfatS.zip this has a single positive rail identical to the kgsshvpssicfetdual2new in topology and has a bias supply has more mounting options for the pass transistor, 4 holes for mounting to a case and can be used with or without an L shaped bracket board size approx. width 138.7mm height 98.6mm (slightly less height than the fat non S version silkscreen revision 1.71s ------------------------------------------------ Version 1.8 boards ------------------------------------------------- same topology as version 1.7 making this a golden reference board adds a cpc1117n and 600ohm resistor to control high voltage startup the high voltage will not start-up automatically. This negates the need for a external relay for delayed start-up and just requires dc power to be supplied to the cpc1117n (not present in earlier versions) for the high voltage to start-up. if you want the board to power up instantly simply omit the 600ohm resistor and cpc1117n and it will act just like the version 1.7 instant start boards. boards same size as the equivalent version 1.7 pcbs silkscreen 1.8 there does not seem to be a dual version of this board kgsshvpssicfetsinglenewleftSWS.zip this has a single negative rail almost identical to the kgsshvpssicfetdual2new in topology and has a simple low voltage positive and negative regulators based supply board size approx. width 76.4mm height 165.3mm silkscreen revision 1.8 kgsshvpssicfetsinglenewrightSWS.zip this has a single positive rail almost identical to the kgsshvpssicfetdual2new in topology and has a bias supply board size approx. width 76.4mm height 165.3mm silkscreen revision 1.8 kgsshvpssicfetsinglenewleftfatSW.zip this has a single negative rail almost identical to the kgsshvpssicfetdual2new in topology and has a simple low voltage positive and negative regulators based supply board size approx. width 138.7mm height 103.7mm silkscreen revision 1.8 kgsshvpssicfetsinglenewrightfatSW.zip this has a single positive rail almost identical to the kgsshvpssicfetdual2new in topology and bias based supply board size approx. width 138.7mm height 103.7mm silkscreen revision 1.8 kgsshvpssicfetsinglenewleftfatSWS.zip this has a single negative rail almost identical to the kgsshvpssicfetdual2new in topology and has a simple low voltage positive and negative regulators based supply has more mounting options for the pass transistor, 4 holes for mounting to a case and can be used with or without an L shaped bracket board size approx. width 138.7mm height 98.6mm (slightly less height than the fat non S version silkscreen revision 1.8 kgsshvpssicfetsinglenewrightfatSWS.zip this has a single positive rail almost identical to the kgsshvpssicfetdual2new in topology and has a bias supply has more mounting options for the pass transistor, 4 holes for mounting to a case and can be used with or without an L shaped bracket board size approx. width 138.7mm height 98.6mm (slightly less height than the fat non S version silkscreen revision 1.8 Edited December 4, 2023 by jamesmking 9 5
Blueman2 Posted February 24, 2021 Report Posted February 24, 2021 (edited) I applaud this thread!!! On 2/20/2021 at 12:59 AM, jamesmking said: 1. part of the headcase experience is trawling a massive number of posts in one thread e.g. the carbon to find out about the grhv. Having a dedicated thread is seen as cheating. 🙂 2. to build a grhv you need C2M1000170D which are out of stock just about everywhere with little hope of being back in stock anytime soon... this is one of the issues delaying my megatron build. On point 1 that James made, I am very guilty of helping cause this issue. I have mixed comments on the GRHV across threads for Blue Hawaii BJT, Carbon, and a few others as well. Just to make sure that it is almost impossible for anyone to learn from what I did. 😁 On point 2, I have tested 2 alternatives to the C2M1000170D, and both appear to work fine. One of them is pretty new and usually in stock: https://www.mouser.com/ProductDetail/?qs=vHuUswq2%2Bsy%2BS7yfkaWaCA%3D%3D Item on left is the one that is in stock. Item to the right is the alternative that I successfully used as well, but often not in stock. Middle is the standard part which appears to be quite elusive. Edited February 24, 2021 by Blueman2 4
Pars Posted February 24, 2021 Author Report Posted February 24, 2021 I managed to get some C2M1000170D from Rochester Electronics (part of Arrow or Allied?) awhile back, but they are out as well now.
JoaMat Posted February 24, 2021 Report Posted February 24, 2021 @Blueman2 Have you tried any of the alternative devices to C2M000170D in a Carbon amplifier?
starcat Posted February 24, 2021 Report Posted February 24, 2021 (edited) C2M1000170D would be available end of April at Farnell and mid of June at Mouser. I have backordered already a pile. Here to check availability across all distributors. Would be nice to have a BOM next to each of the PSUs James has put together above. I will send some shortly. Edited February 24, 2021 by starcat
Blueman2 Posted February 25, 2021 Report Posted February 25, 2021 (edited) 7 hours ago, JoaMat said: @Blueman2 Have you tried any of the alternative devices to C2M000170D in a Carbon amplifier? I have successfully used the ROHM version (3rd column in the list above) for over a year in the Carbon amp with no issues, but I think that part is also out of stock? Used the one if the first column (which has good availability now) in a GRHV with no issues, but not on a Carbon. But from the specs, it should work as well as the other two. Edited February 25, 2021 by Blueman2
Pars Posted February 25, 2021 Author Report Posted February 25, 2021 (edited) This post from Søren in the Circlotron thread is useful for testing a GRHV. On 7/2/2017 at 10:28 AM, sorenb said: I never had troubles with the HV900's, and haven't had any that blew, so the following is based on what I've found useful troubleshooting/checking the regular Golden Reference HV. Often it can be helpful to work at lower voltages when trouble shooting. Paralleling the high resistor (is it 2x 450kohm in the HV900?) in the probe string to get the circuit in action at a lower output DC. If regulation doesn't work for some reason and things go bad it usually isn't really bad. Parallel the R8(2x 450k) with one ~100k (one across both) will provide some ~100VDC. Initial check: The CCS's Use a lab supply at some ~15-20V, through a DMM at mA: put the (+) at the drain (middle leg) of the 10m90, and the (-) at the bottom of the CSS where R2(100) and R1(4k5) meets ...you should get ~2mA ....if you see much more, try replace the DN2540. Do the same at the other CSS: (+) to drain (middle leg) at the 10m90, and (-) at the bottom (pin 6 at the voltage reference) ...you should get ~1.3mA ...if not the 10m90 is probably damaged. Check the current limiting CSS you added (your drawing) the same way (Kevin suggested 50mA for this) When powering up the supply, hold the variac when you reach some ~40-50VDC output, and check: the pin6 of lt1021 against "ground" (pin4) ....you should see 10V ...if not, the lt1021 has probably gone south Before raising the VARIAC further: Attach a DMM across the 2k at the bottom of the cascoded CSS ... you should see ~0V (when at ~50VDC output) and only before regulation kicks in, the voltage jumps to some ~4V; it happens just before you reach the desired output voltage Attach a DMM across the 3k at the source leg of the output CSS (right most leg) ...should be some steady ~4V EDIT: Per Soren, note that these instructions are for the Circlotron, so differ from a GRHV somewhat. For the initial check, I was getting 0.5mA; Soren said maybe 1mA. For the 2nd CCS, I was getting 1.4mA. These will depend upon your particular DN2540s. The current limiting CSS he notes is not applicable on the GRHV, so ignore that portion. My boards both worked fine. The paralleling of the gain resistors (2 x 390K for a 400V GRHV) worked quite well. I put a 100K in parallel with the series string for a regulated output of ~55Vdc. Make sure you use a variac for this to limit the input voltage. Edited December 1, 2022 by Pars Added info 2
luvdunhill Posted March 3, 2021 Report Posted March 3, 2021 Unrelated. But is the REF02 still relevant these days? Need 5V reference and curious in which case it might be a better choice compared to say the LT1021 or something new you folks are using these days.
Sebtdi Posted May 4, 2021 Report Posted May 4, 2021 On 2/24/2021 at 10:57 AM, Blueman2 said: I applaud this thread!!! On point 1 that James made, I am very guilty of helping cause this issue. I have mixed comments on the GRHV across threads for Blue Hawaii BJT, Carbon, and a few others as well. Just to make sure that it is almost impossible for anyone to learn from what I did. 😁 On point 2, I have tested 2 alternatives to the C2M1000170D, and both appear to work fine. One of them is pretty new and usually in stock: https://www.mouser.com/ProductDetail/?qs=vHuUswq2%2Bsy%2BS7yfkaWaCA%3D%3D Item on left is the one that is in stock. Item to the right is the alternative that I successfully used as well, but often not in stock. Middle is the standard part which appears to be quite elusive. Has anyone tried the newer UnitedSIC yet for the GRHV78/79 regulator build in lieu of the Cree C2M1000170D? They are still out of stock until end of June..🥵 Another option that I have found somewhere on the forum is FQPF8N80C. Anyone tried them? https://www.digikey.ca/en/products/detail/on-semiconductor/FQPF8N80C/1056533
Pars Posted May 4, 2021 Author Report Posted May 4, 2021 On 3/2/2021 at 8:53 PM, luvdunhill said: Unrelated. But is the REF02 still relevant these days? Need 5V reference and curious in which case it might be a better choice compared to say the LT1021 or something new you folks are using these days. Not sure Marc. Kevin was using these on the Dynahi PSUs, but switched to the LT1021 for the GRLV/GRHV PSUs. I think the original Dynalo (such as headamp, etc.) also used REF02s.
luvdunhill Posted May 4, 2021 Report Posted May 4, 2021 1 minute ago, Pars said: Not sure Marc. Kevin was using these on the Dynahi PSUs, but switched to the LT1021 for the GRLV/GRHV PSUs. I think the original Dynalo (such as headamp, etc.) also used REF02s. I ended up using the LT1021 - in haven’t tested that circuit yet, but it’s soldered in waiting to get some juice.
johnwmclean Posted May 8, 2021 Report Posted May 8, 2021 (edited) 13297 pcs of cree C2M1000170D https://www.micro-semiconductor.com/products/Cree-Wolfspeed/C2M1000170D Way too cheap.... but looks legit. Edited May 8, 2021 by johnwmclean 1
mwl168 Posted May 8, 2021 Report Posted May 8, 2021 11 hours ago, johnwmclean said: 13297 pcs of cree C2M1000170D https://www.micro-semiconductor.com/products/Cree-Wolfspeed/C2M1000170D Way too cheap.... but looks legit. PM me if anyone in US is interested in sticking your neck out and give this a try. Keep in mind what actually ships may not be what’s in the photo and you can lose every penny invested.
starcat Posted May 11, 2021 Report Posted May 11, 2021 On 2/20/2021 at 5:43 PM, Kerry said: Hi @Kerry, which series exactly is the white board mounted 8-pin 2-row connector in the upper left corner, is this a Molex Mega-Fit? Thanks much again.
Kerry Posted May 11, 2021 Report Posted May 11, 2021 They are the Phoenix Mate-N-Lok-2 connectors. They are rated at 600V which is nice. 1
starcat Posted May 13, 2021 Report Posted May 13, 2021 Also 5.08 pitch which makes it possible to use them instead the usual phoenix block type connectors. Thanks, Kerry.
waltzingbear Posted May 13, 2021 Report Posted May 13, 2021 looking for set of Kerry's psu boards that use the GR78/79 and HV boards. Bare boards are fine with me. TIA cheers Alan
mwl168 Posted May 13, 2021 Report Posted May 13, 2021 1 hour ago, waltzingbear said: looking for set of Kerry's psu boards that use the GR78/79 and HV boards. Bare boards are fine with me. TIA cheers Alan PM me. have both bare and assembled boards.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now