@xxx.edu Fri Jan 6 12:23:47 1995 Date: Fri, 6 Jan 95 09:15:36 GMT @xxx.edu @xxx.za Subject: Majordomo file: list 'diy_efi' file 'archive_num_21' -- >From owner-diy_efi Wed Jul 20 21:00:55 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA21422; Wed, 20 Jul 94 21:00:55 GMT Received: from hwking.cca.rockwell.com by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA21417; Wed, 20 Jul 94 17:00:52 -0400 Received: by hwking.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA22718; Wed, 20 Jul 1994 16:00:14 -0500 Received: by ohura.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA03833; Wed, 20 Jul 1994 16:00:11 -0500 @xxx.com> To: DIY_EFI Subject: Re: Knock sensor In-Reply-To: Your message of "Wed, 20 Jul 94 13:39:35 CDT." @xxx.com> Date: Wed, 20 Jul 94 16:00:10 -0500 @xxx.com X-Mts: smtp Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI >Don't know what range, but I've been told 5-8 Khz. I was gonna notch-filter the thing for 5-15 khz. >I think you will need a low noise op-amp since the signal is low level. >hang on, I've got that article here somewhere. The article shows a scope >hooked to a knock sensor, the display indicates +-5mv straight from the >sensor. One of my co-workers just brought a copy of PE to me. It looks like 500mV, is that what you meant?. I'm thinking the real mccoy will be somewhat higher than that, but it's a good place to start. > sounds like a good idea, let us know if it works out.... It's a must apparently in a Syclone, especially when running high boost levels. I might try to implement it with a LED bar graph, that way I can kinda see what's going on dynamically. I won't get a chance to test the circuit until Labor Day, unfortunately. Thanks for the repost, the info helped a lot. Dig @xxx.com >From owner-diy_efi Thu Jul 21 14:19:19 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23424; Thu, 21 Jul 94 14:19:19 GMT Received: from mbox.ualr.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23419; Thu, 21 Jul 94 10:19:15 -0400 Received: from vn-gateway by UALR.EDU (PMDF V4.2-11 #4503) id @xxx.EDU>; Thu, 21 Jul 1994 09:14:54 CDT Date: Thu, 21 Jul 1994 06:35:00 +0000 @xxx.us (Dave Williams) Subject: Re: Knock sensor To: DIY_EFI @xxx.us> Organization: The Courts of Chaos * 501-985-0059 * Public Access Usenet @xxx.edu Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI -> I was gonna notch-filter the thing for 5-15 khz. SAAB's Trionic engine management system drops voltage after the initial spark, then looks at the resistance of the plug gap. SAAB claims they can tell if the cylinder is running into detonation by looking at this value. If there's an SAE or IMechE paper on this anywhere it'd be interesting. It would save trying to find a place to mount and adjust a sensor. Dunno if the electrical/software part would be harder or not. >From owner-diy_efi Thu Jul 21 14:51:10 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23464; Thu, 21 Jul 94 14:51:10 GMT Received: from oasys.dt.navy.mil by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23459; Thu, 21 Jul 94 10:51:07 -0400 Received: from dtnet7-95.dt.navy.mil by oasys.dt.navy.mil (5.61/oasys.dt.navy.mil) id AA04828; Thu, 21 Jul 94 10:51:05 EDT Date: Thu, 21 Jul 94 10:51:05 EDT @xxx.mil> @xxx.mil Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: DIY_EFI @xxx.mil (Pete Paraska) Subject: Re: Knock sensor X-Mailer: Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Dave Williams wrote: > SAAB's Trionic engine management system drops voltage after the initial >spark, then looks at the resistance of the plug gap. SAAB claims they >can tell if the cylinder is running into detonation by looking at this >value. Just interested in other's ideas how this might actually work. How could plug gap resistance after the spark tell you anything about detonation? Can the plug gap "taste" the residue of preignition as being different from the "taste" (er..resistance) of a slowly propagting flame front that results from correct ignition? It's an interesting concept, though, using plug gap resistance to measure detonation. If it was repeatable and reliable, it would seem to be better than trying to guess what that accelerometer (knock sensor) was REALLY telling you. It seems that all this filtering, amplifying, etc. and making assumptions about what frequency is actually indicative of detonation involves several leaps of faith. @xxx.mil) IZCC#15 ~~~~~~~~~ >From owner-diy_efi Thu Jul 21 15:22:35 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23591; Thu, 21 Jul 94 15:22:35 GMT Received: from hwking.cca.rockwell.com by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23586; Thu, 21 Jul 94 11:22:31 -0400 Received: by hwking.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA24938; Thu, 21 Jul 1994 10:22:28 -0500 Received: by vulcan.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA04625; Thu, 21 Jul 1994 10:22:27 -0500 @xxx.com> To: DIY_EFI Subject: Re: Knock sensor In-Reply-To: Your message of "Thu, 21 Jul 94 10:51:05 EDT." @xxx.mil> Date: Thu, 21 Jul 94 10:22:27 -0500 @xxx.com X-Mts: smtp Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI >Just interested in other's ideas how this might actually work. How could >plug gap resistance after the spark tell you anything about detonation? Can >the plug gap "taste" the residue of preignition as being different from the >"taste" (er..resistance) of a slowly propagting flame front that results >from correct ignition? Kenne-Bell is marketing an ignition controller for Syclones/Typhoons that can supposedly detect detonation in each cylinder and retards timing only in that cylinder. Since they don't use a knock sensor for each cyl., this must be how they determine whether or not a cylinder is detonating. Could it be that they are considering unusually high cylinder pressure to be equal to detonation? The resistance of the gap would go up, the current would drop- there's your detection. Sounds a lot like a Jacobs ignition. Dig @xxx.com >From owner-diy_efi Thu Jul 21 16:21:49 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23790; Thu, 21 Jul 94 16:21:49 GMT Received: from mn.ecn.purdue.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23779; Thu, 21 Jul 94 12:21:11 -0400 Received: from mn.ecn.purdue.edu (armfield@localhost) by mn.ecn.purdue.edu (8.6.8/3.5davy) id LAA04744; Thu, 21 Jul 1994 11:20:45 -0500 @xxx.edu> Date: Thu, 21 Jul 1994 11:20:45 -0500 @xxx.edu> To: DIY_EFI Subject: Re : Knock sensor - Kenne-Bell System Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI I would doubt (IMHO) that Kenne Bell is using anything as sophisticated as looking at resistance measurements to do individual cylinder knock control. My guess is that they only "listen" to the knock sensor during a period a little before and a little after TDC on the firing stroke of each cylinder. Then, theoretically, they would only "hear" the knock of the cylinder firing. After you have established which cylinder is firing and if it is indeed knocking, then changing the timing on a cylinder by cylinder basis is no big deal anymore. Jeff Armfield >From owner-diy_efi Thu Jul 21 17:00:00 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23911; Thu, 21 Jul 94 17:00:00 GMT Received: from nic.hookup.net by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23906; Thu, 21 Jul 94 12:59:52 -0400 @xxx.206) with UUCP id MAA19919; Thu, 21 Jul 1994 12:55:25 -0400 Received: by ndigital.com (1.64/waf) via UUCP; Thu, 21 Jul 94 12:24:26 EDT @xxx.edu @xxx.com (Mike Palmer) @xxx.com> Date: Thu, 21 Jul 94 12:24:25 EDT Organization: Northern Digital Inc. Waterloo, Ontario, Canada To: DIY_EFI Subject: Re: Knock... Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI >>Just interested in other's ideas how this might actually work. How could >>plug gap resistance after the spark tell you anything about detonation? Can >>the plug gap "taste" the residue of preignition as being different from the >>"taste" (er..resistance) of a slowly propagting flame front that results >>from correct ignition? > >Kenne-Bell is marketing an ignition controller for Syclones/Typhoons that >can supposedly detect detonation in each cylinder and retards timing only >in that cylinder. Since they don't use a knock sensor for each cyl., this >must be how they determine whether or not a cylinder is detonating. > >Could it be that they are considering unusually high cylinder pressure to >be equal to detonation? The resistance of the gap would go up, the current >would drop- there's your detection. Sounds a lot like a Jacobs ignition. > >Dig @xxx.com Hmmm. Since detonation occurs generally a goodly amount of time after the primary spark has ignited the mixture, in order to determine a gap resistance value they'd have to keep peppering the plug with sparks and monitoring firing voltage. If this is the case, then how do they get sufficient coil saturation between "check-sparks" to induce enough secondary to actually fire the plug? If they don't measure gap resistance this way, then how exactly do they do it (i.e. measuring the effective gap resistance over the course of the combustion period to detect unusual pressure gradients indicating abnormal combustion)? Just curious... -- - MJP >From owner-diy_efi Thu Jul 21 17:12:56 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA23954; Thu, 21 Jul 94 17:12:56 GMT Received: from hwking.cca.rockwell.com by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA23949; Thu, 21 Jul 94 13:12:52 -0400 Received: by hwking.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA26209; Thu, 21 Jul 1994 12:12:43 -0500 Received: by vulcan.cca.rockwell.com (5.65/DEC-Ultrix/4.3) id AA04727; Thu, 21 Jul 1994 12:12:43 -0500 @xxx.com> To: DIY_EFI Subject: Re: Re : Knock sensor - Kenne-Bell System In-Reply-To: Your message of "Thu, 21 Jul 94 11:20:45 CDT." @xxx.edu> Date: Thu, 21 Jul 94 12:12:42 -0500 @xxx.com X-Mts: smtp Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI >I would doubt (IMHO) that Kenne Bell is using anything as sophisticated >as looking at resistance measurements to do individual cylinder knock >control. If KB had built the thing themselves, I would agree. It was built by some outfit out west, Jones electronics? maybe. It came out about the same time as the Jacobs unit, which made me suspect that they used similar technology. >My guess is that they only "listen" to the knock sensor during a >period a little before and a little after TDC on the firing stroke of each >cylinder. I thought about that. Could that work with a distributor ignition without a cam or crank position sensor? Dig @xxx.com >From owner-diy_efi Thu Jul 21 17:32:12 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA24019; Thu, 21 Jul 94 17:32:12 GMT Received: from knuth.mtsu.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA24014; Thu, 21 Jul 94 13:32:02 -0400 Received: by knuth.mtsu.edu (Smail3.1.28.1 #17) id m0qQzoz-000AXZC; Thu, 21 Jul 94 10:14 CDT @xxx.edu> @xxx. Lusky) Subject: Re: Knock sensor To: DIY_EFI Date: Thu, 21 Jul 1994 10:14:05 -0500 (CDT) @xxx.mil> from "Pete Paraska" at Jul 21, 94 10:51:05 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 337 Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Pete Paraska writes: > than trying to guess what that accelerometer (knock sensor) was REALLY > telling you. Huh? A knock sensor is just a microphone basically... -- @xxx.edu "Turbos are nice but I'd rather be blown!" 68 Camaro Convertible - 350 / TH350 80 Toyota Celica - 20R / 5spd >From owner-diy_efi Fri Jul 22 15:31:38 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA26678; Fri, 22 Jul 94 15:31:38 GMT Received: from mbox.ualr.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA26672; Fri, 22 Jul 94 11:31:34 -0400 Received: from vn-gateway by UALR.EDU (PMDF V4.2-11 #4503) id @xxx.EDU>; Fri, 22 Jul 1994 10:31:09 CDT Date: Fri, 22 Jul 1994 03:49:00 +0000 @xxx.us (Dave Williams) Subject: Re: Knock sensor To: DIY_EFI @xxx.us> Organization: The Courts of Chaos * 501-985-0059 * Public Access Usenet @xxx.edu Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI -> It seems that all this filtering, amplifying, etc. and making -> assumptions about what frequency is actually indicative of detonation -> involves several leaps of faith. Cosworth doesn't (or at least didn't up to a couple of years ago) use knock sensors in their F1 motors. Keith Duckworth said the engines were too noisy for the signal to mean anything. The popular TAG management system used my many F1 teams can accept several knock sensors, but if other engines are like the Cosworth V8s the sensors might be just data acquisition in most cases. BTW, TAG claims their newer management systems vary spark energy according to load and RPM, though I'm not sure what advantage this would be. Shades of Dr. Jacobs and his miracle boxes... >From owner-diy_efi Fri Jul 22 15:31:35 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA26673; Fri, 22 Jul 94 15:31:35 GMT Received: from mbox.ualr.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA26667; Fri, 22 Jul 94 11:31:28 -0400 Received: from vn-gateway by UALR.EDU (PMDF V4.2-11 #4503) id @xxx.EDU>; Fri, 22 Jul 1994 10:30:59 CDT Date: Fri, 22 Jul 1994 03:44:00 +0000 @xxx.us (Dave Williams) Subject: Re: Knock sensor To: DIY_EFI @xxx.us> Organization: The Courts of Chaos * 501-985-0059 * Public Access Usenet @xxx.edu Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI -> Just interested in other's ideas how this might actually work. How -> could plug gap resistance after the spark tell you anything about -> detonation? As chamber pressure goes up it becomes harder for a spark to jump the gap. SAAB might be doing that... but you'd actually have to send a spark, which wouldn't be a good idea before the intended ignition event. They're probably watching for ionization from detonation combustion. The resistance across the gap will change if the mixture is burning. >From owner-diy_efi Fri Jul 22 15:32:02 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA26686; Fri, 22 Jul 94 15:32:02 GMT Received: from mbox.ualr.edu by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA26681; Fri, 22 Jul 94 11:32:00 -0400 Received: from vn-gateway by UALR.EDU (PMDF V4.2-11 #4503) id @xxx.EDU>; Fri, 22 Jul 1994 10:31:23 CDT Date: Fri, 22 Jul 1994 03:53:00 +0000 @xxx.us (Dave Williams) Subject: Re: Knock sensor To: DIY_EFI @xxx.us> Organization: The Courts of Chaos * 501-985-0059 * Public Access Usenet @xxx.edu Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI -> Kenne-Bell is marketing an ignition controller for Syclones/Typhoons -> that can supposedly detect detonation in each cylinder and retards -> timing only in that cylinder. Since they don't use a knock sensor for -> each cyl., this must be how they determine whether or not a cylinder -> is detonating. Lots of engines do cylinder-by-cylinder with only one or two knock sensors. You know, either by crank position or firing order, which cylinder is coming up on the compression stroke. If you get a knock signal, it's gonna be that cylinder. >From owner-diy_efi Mon Jul 25 11:38:56 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA04067; Mon, 25 Jul 94 11:38:56 GMT Received: from oasys.dt.navy.mil by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA04062; Mon, 25 Jul 94 07:38:53 -0400 Received: from dtnet7-95.dt.navy.mil by oasys.dt.navy.mil (5.61/oasys.dt.navy.mil) id AA11173; Mon, 25 Jul 94 07:38:50 EDT Date: Mon, 25 Jul 94 07:38:50 EDT @xxx.mil> @xxx.mil Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: DIY_EFI @xxx.mil (Pete Paraska) Subject: Re: Knock sensor X-Mailer: Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI @xxx.us (Dave Williams) replied: > As chamber pressure goes up it becomes harder for a spark to jump the >gap. SAAB might be doing that... but you'd actually have to send a >spark, which wouldn't be a good idea before the intended ignition event. >They're probably watching for ionization from detonation combustion. >The resistance across the gap will change if the mixture is burning. > I just read in Automotive Industries (AI) annual engine report that Dave is right on the money - they are keying to ionization. They didn't mention how they where getting a reading on that, but Dave cleared that up - resistance. The system is call Trionic Ignition or some such. Thanks (again) Dave, for giving us the right stuff. @xxx.mil) IZCC#15 ~~~~~~~~~ >From owner-diy_efi Wed Aug 3 13:03:49 1994 Received: by coulomb.eng.ohio-state.edu (920330.SGI/920502.SGI) id AA04282; Wed, 3 Aug 94 13:03:49 GMT Received: from ra.ibr.cs.tu-bs.de by coulomb.eng.ohio-state.edu via SMTP (920330.SGI/920502.SGI) for /usr/local/mail/majordomo-1.92/wrapper resend -p bulk -M 10000 -l Diy_Efi -f Diy_Efi-Owner -h coulomb.eng.ohio-state.edu -s -r DIY_EFI diy_efi-outgoing id AA04270; Wed, 3 Aug 94 09:03:25 -0400 Received: from diana by ra.ibr.cs.tu-bs.de (5.65/1.341) id AA08900; Wed, 3 Aug 94 15:03:09 +0200 Received: from bal by diana (4.1/SMI-4.1N) id AA18864; Wed, 3 Aug 94 15:03:07 +0200 Date: Wed, 3 Aug 94 15:03:06 +0200 @xxx.de (Jens Knickmeyer) Message-Id: <9408031303.AA18864@diana> To: DIY_EFI Subject: VW Digifant Sender: owner-diy_efi Precedence: bulk Reply-To: DIY_EFI Being new on this list, I would like to give a short introduction of myself: I am a 26 year old German student of computer science. My hobbies are electrical engeneering and cars, especially VW-types. I work at the mircoprocessor laboratory of my university and therefore I have the chance to get information about Motorola microprocessors and controllers. Last, I would like to apologize my rather bad English. So far for my person, now to the real things: Since I drive a VW Polo-G40 I am interested in getting any information about the EFI used in the Polo-G40 (BTW, the same EFI is used in the Golf/Corrado/Passat G60!), the "VW Digifant" which is based on the Bosch motronic (multipoint FI, using a 68HC11 microcontroller). Especially, need answers for the following questions: - There is a build-in self diagnostic system in the Digifant which uses only one wire to send and receive information. 1. Which are the electrical characteristics of this interface? 2. Is it the "ABUS"-interface developped by VW and Intel/Telefunken? 3. What protocol is used? - The Digifant stores injection/ignition data in a 27C512 EPROM chip. 1. How are these data organized? 2. How do I have to change EPROM data after tuning the mechanical G-compressor used in the G-40/G-60 ? Thanks in advance, Jens. \|/ o o -----------------o0--(_)--0o------------------- @xxx.de ÿ